modelId
stringlengths 5
139
| author
stringlengths 2
42
| last_modified
timestamp[us, tz=UTC]date 2020-02-15 11:33:14
2025-09-03 00:36:49
| downloads
int64 0
223M
| likes
int64 0
11.7k
| library_name
stringclasses 535
values | tags
listlengths 1
4.05k
| pipeline_tag
stringclasses 55
values | createdAt
timestamp[us, tz=UTC]date 2022-03-02 23:29:04
2025-09-03 00:36:49
| card
stringlengths 11
1.01M
|
|---|---|---|---|---|---|---|---|---|---|
aga3134/ppo-SnowballTarget
|
aga3134
| 2023-06-13T03:20:04Z | 28 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"SnowballTarget",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-SnowballTarget",
"region:us"
] |
reinforcement-learning
| 2023-06-13T03:20:01Z |
---
library_name: ml-agents
tags:
- SnowballTarget
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-SnowballTarget
---
# **ppo** Agent playing **SnowballTarget**
This is a trained model of a **ppo** agent playing **SnowballTarget**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: aga3134/ppo-SnowballTarget
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
koki0702/test-butterflies-32
|
koki0702
| 2023-06-13T01:50:24Z | 30 | 0 |
diffusers
|
[
"diffusers",
"pytorch",
"unconditional-image-generation",
"diffusion-models-class",
"license:mit",
"diffusers:DDPMPipeline",
"region:us"
] |
unconditional-image-generation
| 2023-06-13T01:49:51Z |
---
license: mit
tags:
- pytorch
- diffusers
- unconditional-image-generation
- diffusion-models-class
---
# Model Card for Unit 1 of the [Diffusion Models Class 🧨](https://github.com/huggingface/diffusion-models-class)
This model is a diffusion model for unconditional image generation of cute 🦋.
## Usage
```python
from diffusers import DDPMPipeline
pipeline = DDPMPipeline.from_pretrained('koki0702/test-butterflies-32')
image = pipeline().images[0]
image
```
|
yyabuki/pegasus-samsum
|
yyabuki
| 2023-06-13T01:35:21Z | 96 | 0 |
transformers
|
[
"transformers",
"pytorch",
"pegasus",
"text2text-generation",
"generated_from_trainer",
"dataset:samsum",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2023-06-12T16:06:37Z |
---
tags:
- generated_from_trainer
datasets:
- samsum
model-index:
- name: pegasus-samsum
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# pegasus-samsum
This model is a fine-tuned version of [google/pegasus-cnn_dailymail](https://huggingface.co/google/pegasus-cnn_dailymail) on the samsum dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 1
### Framework versions
- Transformers 4.28.1
- Pytorch 1.10.2
- Datasets 2.11.0
- Tokenizers 0.13.3
|
FremyCompany/opus-mt-nl-en-healthcare
|
FremyCompany
| 2023-06-13T01:31:02Z | 131 | 2 |
transformers
|
[
"transformers",
"pytorch",
"safetensors",
"marian",
"text2text-generation",
"translation",
"nl",
"en",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
translation
| 2023-04-17T06:44:40Z |
---
license: mit
language:
- nl
- en
pipeline_tag: translation
---
# Opus-MT-NL-EN-Healthcare
This model can be used to translate medical text from Dutch to English, as a baseline for Clinical NLP systems in Dutch.
## Model Details
### Model Description
This model based on MarianMT was trained using OpusNLP data as well as specialized medical data (see paper for more details).
- **Developed by:** François Remy
- **Model type:** Translation
- **Languages:** From Dutch to English
- **License:** MIT
- **Finetuned from model:** Helsinki-NLP/opus-mt-nl-en
### Model Sources
- **Repository:** [Snomed Translation Dictionaries](https://github.com/FremyCompany/snomed-translate-dictionaries)
- **Paper:** [Taming large lexicons: translating clinical text using medical ontologies and sentence templates](https://biblio.ugent.be/publication/01GMN7J3X37CKWCKQV6DWGM4ZE/file/01GMN7NT30DKF37RXN31DSJZ2D.pdf)
## Citation
If you use this model as part of your work, please cite the following paper:
**BibTeX:**
```
@inproceedings{remy-et-al-2022-translating,
author = {{Remy, François and De Jaeger, P. and Demuynck, Kris}},
title = {{Taming large lexicons: translating clinical text using medical ontologies and sentence templates}},
booktitle = {{EmP 2022}}
location = {{Roeselare, Belgium}},
year = {{2022}},
language = {{en}},
pages = {{5}},
}
```
**APA:**
```
Remy, F., De Jaeger, P., & Demuynck, K. (2022). Taming large lexicons: translating clinical text using medical ontologies and sentence templates. Presented at the EmP 2022, The “Engineers meet Practicians” conference, Roeselare, Belgium.
```
|
withxiao/alpaca-llama-7b-fp16
|
withxiao
| 2023-06-13T01:27:18Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:yahma/alpaca-cleaned",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-10T05:00:41Z |
---
datasets:
- yahma/alpaca-cleaned
pipeline_tag: text-generation
---
|
GoGD/xlm-roberta-base-finetuned-panx-all
|
GoGD
| 2023-06-13T01:25:21Z | 101 | 0 |
transformers
|
[
"transformers",
"pytorch",
"xlm-roberta",
"token-classification",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
token-classification
| 2023-06-13T01:20:24Z |
---
license: mit
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: xlm-roberta-base-finetuned-panx-all
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# xlm-roberta-base-finetuned-panx-all
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1297
- F1: 0.8803
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 24
- eval_batch_size: 24
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 0.2887 | 1.0 | 620 | 0.1770 | 0.8207 |
| 0.1426 | 2.0 | 1240 | 0.1468 | 0.8578 |
| 0.091 | 3.0 | 1860 | 0.1297 | 0.8803 |
### Framework versions
- Transformers 4.16.2
- Pytorch 1.12.1
- Datasets 1.16.1
- Tokenizers 0.13.3
|
timdettmers/qlora-unnatural-instructions-65b
|
timdettmers
| 2023-06-13T01:24:44Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T19:15:01Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-self-instruct-65b
|
timdettmers
| 2023-06-13T01:24:34Z | 0 | 2 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T20:20:08Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-alpaca-65b
|
timdettmers
| 2023-06-13T01:24:30Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T19:13:42Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-chip2-65b
|
timdettmers
| 2023-06-13T01:24:26Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T20:31:11Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-flan-65b
|
timdettmers
| 2023-06-13T01:24:21Z | 0 | 1 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T19:14:00Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-hh-rlhf-33b
|
timdettmers
| 2023-06-13T01:24:14Z | 0 | 1 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T21:33:20Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-self-instruct-33b
|
timdettmers
| 2023-06-13T01:24:11Z | 0 | 4 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T21:17:03Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-longform-33b
|
timdettmers
| 2023-06-13T01:24:07Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T19:07:07Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-flan-33b
|
timdettmers
| 2023-06-13T01:23:59Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T20:36:41Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-unnatural-instructions-13b
|
timdettmers
| 2023-06-13T01:23:55Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T19:01:03Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-self-instruct-13b
|
timdettmers
| 2023-06-13T01:23:50Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T18:55:40Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-longform-13b
|
timdettmers
| 2023-06-13T01:23:47Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T18:56:13Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-chip2-13b
|
timdettmers
| 2023-06-13T01:23:45Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T18:55:31Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-flan-13b
|
timdettmers
| 2023-06-13T01:23:41Z | 0 | 0 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T19:02:40Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/qlora-alpaca-7b
|
timdettmers
| 2023-06-13T01:23:27Z | 0 | 3 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"region:us"
] | null | 2023-05-22T18:48:52Z |
# QLoRA Instruction Tuned Models
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The `QLoRA Instruction Tuned Models` are open-source models obtained through 4-bit QLoRA tuning of LLaMA base models on various instruction tuning datasets. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
**Note: The best performing chatbot models are named [Guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco) and finetuned on OASST1. This model card is for the other models finetuned on other instruction tuning datasets.**
⚠️ These models are purely intended for research purposes and could produce problematic outputs.
## What are QLoRA Instruction Tuned Models and why use them?
- **Strong performance on MMLU** following the QLoRA instruction tuning.
- **Replicable and efficient instruction tuning procedure** that can be extended to new use cases. QLoRA training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
QLoRA Instruction Tuned adapter weights are available under Apache 2 license. Note the use of these adapter weights, requires access to the LLaMA model weighs and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Flan v2 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/qlora-flan-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The models released here are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: These models use LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that these models can inherit biases and limitations of the base model.
**Finetuning Data**: These models are finetuned on various instruction tuning datasets. The datasets used are: Alpaca, HH-RLHF, Unnatural Instr., Chip2, Longform, Self-Instruct, FLAN v2.
**Languages**: The different datasets cover different languages. We direct to the various papers and resources describing the datasets for more details.
Next, we describe Training and Evaluation details.
### Training
QLoRA Instruction Tuned Models are the result of 4-bit QLoRA supervised finetuning on different instruction tuning datasets.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
| Parameters | Dataset | Batch size | LR | Steps | Source Length | Target Length |
|------------|----------|------------|------|-------|---------------|---------------|
| 7B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 7B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 7B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 7B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 13B | All | 16 | 2e-4 | 10000 | 384 | 128 |
| 13B | OASST1 | 16 | 2e-4 | 1875 | - | 512 |
| 13B | HH-RLHF | 16 | 2e-4 | 10000 | - | 768 |
| 13B | Longform | 16 | 2e-4 | 4000 | 512 | 1024 |
| 33B | All | 32 | 1e-4 | 5000 | 384 | 128 |
| 33B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 33B | HH-RLHF | 32 | 1e-4 | 5000 | - | 768 |
| 33B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
| 65B | All | 64 | 1e-4 | 2500 | 384 | 128 |
| 65B | OASST1 | 16 | 1e-4 | 1875 | - | 512 |
| 65B | HH-RLHF | 64 | 1e-4 | 2500 | - | 768 |
| 65B | Longform | 32 | 1e-4 | 2343 | 512 | 1024 |
### Evaluation
We use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
We evaluate the generative language capabilities through automated evaluations on the Vicuna benchmark. We report the score of the QLoRA Instruction Finetuned Models relative to the score obtained by ChatGPT. The rater in this case is GPT-4 which is tasked to assign a score out of 10 to both ChatGPT and the model outputs for each prompt. We report scores for models ranging 7B to 65B and compare them to both academic and commercial baselilnes.
| Model / Dataset | Params | Model bits | Memory | ChatGPT vs Sys | Sys vs ChatGPT | Mean | 95\% CI |
|------------------|--------|------------|--------|----------------|----------------|------------------|---------|
| GPT-4 | - | - | - | 119.4\% | 110.1\% | **114.5**\% | 2.6\% |
| Bard | - | - | - | 93.2\% | 96.4\% | 94.8\% | 4.1\% |
| Guanaco | 65B | 4-bit | 41 GB | 96.7\% | 101.9\% | **99.3**\% | 4.4\% |
| Alpaca | 65B | 4-bit | 41 GB | 63.0\% | 77.9\% | 70.7\% | 4.3\% |
| FLAN v2 | 65B | 4-bit | 41 GB | 37.0\% | 59.6\% | 48.4\% | 4.6\% |
| Guanaco | 33B | 4-bit | 21 GB | 96.5\% | 99.2\% | **97.8**\% | 4.4\% |
| Open Assistant | 33B | 16-bit | 66 GB | 73.4\% | 85.7\% | 78.1\% | 5.3\% |
| Alpaca | 33B | 4-bit | 21 GB | 67.2\% | 79.7\% | 73.6\% | 4.2\% |
| FLAN v2 | 33B | 4-bit | 21 GB | 26.3\% | 49.7\% | 38.0\% | 3.9\% |
| Vicuna | 13B | 16-bit | 26 GB | 91.2\% | 98.7\% | **94.9**\% | 4.5\% |
| Guanaco | 13B | 4-bit | 10 GB | 87.3\% | 93.4\% | 90.4\% | 5.2\% |
| Alpaca | 13B | 4-bit | 10 GB | 63.8\% | 76.7\% | 69.4\% | 4.2\% |
| HH-RLHF | 13B | 4-bit | 10 GB | 55.5\% | 69.1\% | 62.5\% | 4.7\% |
| Unnatural Instr. | 13B | 4-bit | 10 GB | 50.6\% | 69.8\% | 60.5\% | 4.2\% |
| Chip2 | 13B | 4-bit | 10 GB | 49.2\% | 69.3\% | 59.5\% | 4.7\% |
| Longform | 13B | 4-bit | 10 GB | 44.9\% | 62.0\% | 53.6\% | 5.2\% |
| Self-Instruct | 13B | 4-bit | 10 GB | 38.0\% | 60.5\% | 49.1\% | 4.6\% |
| FLAN v2 | 13B | 4-bit | 10 GB | 32.4\% | 61.2\% | 47.0\% | 3.6\% |
| Guanaco | 7B | 4-bit | 5 GB | 84.1\% | 89.8\% | **87.0**\% | 5.4\% |
| Alpaca | 7B | 4-bit | 5 GB | 57.3\% | 71.2\% | 64.4\% | 5.0\% |
| FLAN v2 | 7B | 4-bit | 5 GB | 33.3\% | 56.1\% | 44.8\% | 4.0\% |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/guanaco-33b
|
timdettmers
| 2023-06-13T01:21:26Z | 0 | 27 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"arxiv:2304.07327",
"region:us"
] | null | 2023-05-22T18:18:54Z |
# Guanaco Models Based on LLaMA
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The Guanaco models are open-source finetuned chatbots obtained through 4-bit QLoRA tuning of LLaMA base models on the OASST1 dataset. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
⚠️Guanaco is a model purely intended for research purposes and could produce problematic outputs.
## Why use Guanaco?
- **Competitive with commercial chatbot systems on the Vicuna and OpenAssistant benchmarks** (ChatGPT and BARD) according to human and GPT-4 raters. We note that the relative performance on tasks not covered in these benchmarks could be very different. In addition, commercial systems evolve over time (we used outputs from the March 2023 version of the models).
- **Available open-source for research purposes**. Guanaco models allow *cheap* and *local* experimentation with high-quality chatbot systems.
- **Replicable and efficient training procedure** that can be extended to new use cases. Guanaco training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
Guanaco adapter weights are available under Apache 2 license. Note the use of the Guanaco adapter weights, requires access to the LLaMA model weighs.
Guanaco is based on LLaMA and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Guanaco 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/guanaco-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/guanaco-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The Guanaco models are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: Guanaco uses LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that Guanaco can inherit biases and limitations of the base model.
**Finetuning Data**: Guanaco is finetuned on OASST1. The exact dataset is available at [timdettmers/openassistant-guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco).
**Languages**: The OASST1 dataset is multilingual (see [the paper](https://arxiv.org/abs/2304.07327) for details) and as such Guanaco responds to user queries in different languages. We note, however, that OASST1 is heavy in high-resource languages. In addition, human evaluation of Guanaco was only performed in English and based on qualitative analysis we observed degradation in performance in other languages.
Next, we describe Training and Evaluation details.
### Training
Guanaco models are the result of 4-bit QLoRA supervised finetuning on the OASST1 dataset.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
Size| Dataset | Batch Size | Learning Rate | Max Steps | Sequence length
---|---|---|---|---|---
7B | OASST1 | 16 | 2e-4 | 1875 | 512
13B | OASST1 | 16 | 2e-4 | 1875 | 512
33B | OASST1 | 16 | 1e-4 | 1875 | 512
65B | OASST1 | 16 | 1e-4 | 1875 | 512
### Evaluation
We test generative language capabilities through both automated and human evaluations. This second set of evaluations relies on queries curated by humans and aims at measuring the quality of model responses. We use the Vicuna and OpenAssistant datasets with 80 and 953 prompts respectively.
In both human and automated evaluations, for each prompt, raters compare all pairs of responses across the models considered. For human raters we randomize the order of the systems, for GPT-4 we evaluate with both orders.
Benchmark | Vicuna | | Vicuna | | OpenAssistant | | -
-----------|----|-----|--------|---|---------------|---|---
Prompts | 80 | | 80 | | 953 | |
Judge | Human | | GPT-4 | | GPT-4 | |
Model | Elo | Rank | Elo | Rank | Elo | Rank | **Median Rank**
GPT-4 | 1176 | 1 | 1348 | 1 | 1294 | 1 | 1
Guanaco-65B | 1023 | 2 | 1022 | 2 | 1008 | 3 | 2
Guanaco-33B | 1009 | 4 | 992 | 3 | 1002 | 4 | 4
ChatGPT-3.5 Turbo | 916 | 7 | 966 | 5 | 1015 | 2 | 5
Vicuna-13B | 984 | 5 | 974 | 4 | 936 | 5 | 5
Guanaco-13B | 975 | 6 | 913 | 6 | 885 | 6 | 6
Guanaco-7B | 1010 | 3 | 879 | 8 | 860 | 7 | 7
Bard | 909 | 8 | 902 | 7 | - | - | 8
We also use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
## Risks and Biases
The model can produce factually incorrect output, and should not be relied on to produce factually accurate information. The model was trained on various public datasets; it is possible that this model could generate lewd, biased, or otherwise offensive outputs.
However, we note that finetuning on OASST1 seems to reduce biases as measured on the CrowS dataset. We report here the performance of Guanaco-65B compared to other baseline models on the CrowS dataset.
| | LLaMA-65B | GPT-3 | OPT-175B | Guanaco-65B |
|----------------------|-----------|-------|----------|---------------|
| Gender | 70.6 | 62.6 | 65.7 | **47.5** |
| Religion | {79.0} | 73.3 | 68.6 | **38.7** |
| Race/Color | 57.0 | 64.7 | 68.6 | **45.3** |
| Sexual orientation | {81.0} | 76.2 | 78.6 | **59.1** |
| Age | 70.1 | 64.4 | 67.8 | **36.3** |
| Nationality | 64.2 | 61.6 | 62.9 | **32.4** |
| Disability | 66.7 | 76.7 | 76.7 | **33.9** |
| Physical appearance | 77.8 | 74.6 | 76.2 | **43.1** |
| Socioeconomic status | 71.5 | 73.8 | 76.2 | **55.3** |
| Average | 66.6 | 67.2 | 69.5 | **43.5** |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
timdettmers/guanaco-13b
|
timdettmers
| 2023-06-13T01:21:14Z | 0 | 18 | null |
[
"arxiv:2305.14314",
"arxiv:2302.13971",
"arxiv:2304.07327",
"region:us"
] | null | 2023-05-22T18:26:50Z |
# Guanaco Models Based on LLaMA
| [Paper](https://arxiv.org/abs/2305.14314) | [Code](https://github.com/artidoro/qlora) | [Demo](https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi) |
**The Guanaco models are open-source finetuned chatbots obtained through 4-bit QLoRA tuning of LLaMA base models on the OASST1 dataset. They are available in 7B, 13B, 33B, and 65B parameter sizes.**
⚠️Guanaco is a model purely intended for research purposes and could produce problematic outputs.
## Why use Guanaco?
- **Competitive with commercial chatbot systems on the Vicuna and OpenAssistant benchmarks** (ChatGPT and BARD) according to human and GPT-4 raters. We note that the relative performance on tasks not covered in these benchmarks could be very different. In addition, commercial systems evolve over time (we used outputs from the March 2023 version of the models).
- **Available open-source for research purposes**. Guanaco models allow *cheap* and *local* experimentation with high-quality chatbot systems.
- **Replicable and efficient training procedure** that can be extended to new use cases. Guanaco training scripts are available in the [QLoRA repo](https://github.com/artidoro/qlora).
- **Rigorous comparison to 16-bit methods** (both 16-bit full-finetuning and LoRA) in [our paper](https://arxiv.org/abs/2305.14314) demonstrates the effectiveness of 4-bit QLoRA finetuning.
- **Lightweight** checkpoints which only contain adapter weights.
## License and Intended Use
Guanaco adapter weights are available under Apache 2 license. Note the use of the Guanaco adapter weights, requires access to the LLaMA model weighs.
Guanaco is based on LLaMA and therefore should be used according to the LLaMA license.
## Usage
Here is an example of how you would load Guanaco 7B in 4-bits:
```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/guanaco-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
load_in_4bit=True,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type='nf4'
),
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Inference can then be performed as usual with HF models as follows:
```python
prompt = "Introduce yourself"
formatted_prompt = (
f"A chat between a curious human and an artificial intelligence assistant."
f"The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
f"### Human: {prompt} ### Assistant:"
)
inputs = tokenizer(formatted_prompt, return_tensors="pt").to("cuda:0")
outputs = model.generate(inputs=inputs.input_ids, max_new_tokens=20)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```
Expected output similar to the following:
```
A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions.
### Human: Introduce yourself ### Assistant: I am an artificial intelligence assistant. I am here to help you with any questions you may have.
```
## Current Inference Limitations
Currently, 4-bit inference is slow. We recommend loading in 16 bits if inference speed is a concern. We are actively working on releasing efficient 4-bit inference kernels.
Below is how you would load the model in 16 bits:
```python
model_name = "huggyllama/llama-7b"
adapters_name = 'timdettmers/guanaco-7b'
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.bfloat16,
device_map="auto",
max_memory= {i: '24000MB' for i in range(torch.cuda.device_count())},
)
model = PeftModel.from_pretrained(model, adapters_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
```
## Model Card
**Architecture**: The Guanaco models are LoRA adapters to be used on top of LLaMA models. They are added to all layers. For all model sizes, we use $r=64$.
**Base Model**: Guanaco uses LLaMA as base model with sizes 7B, 13B, 33B, 65B. LLaMA is a causal language model pretrained on a large corpus of text. See [LLaMA paper](https://arxiv.org/abs/2302.13971) for more details. Note that Guanaco can inherit biases and limitations of the base model.
**Finetuning Data**: Guanaco is finetuned on OASST1. The exact dataset is available at [timdettmers/openassistant-guanaco](https://huggingface.co/datasets/timdettmers/openassistant-guanaco).
**Languages**: The OASST1 dataset is multilingual (see [the paper](https://arxiv.org/abs/2304.07327) for details) and as such Guanaco responds to user queries in different languages. We note, however, that OASST1 is heavy in high-resource languages. In addition, human evaluation of Guanaco was only performed in English and based on qualitative analysis we observed degradation in performance in other languages.
Next, we describe Training and Evaluation details.
### Training
Guanaco models are the result of 4-bit QLoRA supervised finetuning on the OASST1 dataset.
All models use NormalFloat4 datatype for the base model and LoRA adapters on all linear layers with BFloat16 as computation datatype. We set LoRA $r=64$, $\alpha=16$. We also use Adam beta2 of 0.999, max grad norm of 0.3 and LoRA dropout of 0.1 for models up to 13B and 0.05 for 33B and 65B models.
For the finetuning process, we use constant learning rate schedule and paged AdamW optimizer.
### Training hyperparameters
Size| Dataset | Batch Size | Learning Rate | Max Steps | Sequence length
---|---|---|---|---|---
7B | OASST1 | 16 | 2e-4 | 1875 | 512
13B | OASST1 | 16 | 2e-4 | 1875 | 512
33B | OASST1 | 16 | 1e-4 | 1875 | 512
65B | OASST1 | 16 | 1e-4 | 1875 | 512
### Evaluation
We test generative language capabilities through both automated and human evaluations. This second set of evaluations relies on queries curated by humans and aims at measuring the quality of model responses. We use the Vicuna and OpenAssistant datasets with 80 and 953 prompts respectively.
In both human and automated evaluations, for each prompt, raters compare all pairs of responses across the models considered. For human raters we randomize the order of the systems, for GPT-4 we evaluate with both orders.
Benchmark | Vicuna | | Vicuna | | OpenAssistant | | -
-----------|----|-----|--------|---|---------------|---|---
Prompts | 80 | | 80 | | 953 | |
Judge | Human | | GPT-4 | | GPT-4 | |
Model | Elo | Rank | Elo | Rank | Elo | Rank | **Median Rank**
GPT-4 | 1176 | 1 | 1348 | 1 | 1294 | 1 | 1
Guanaco-65B | 1023 | 2 | 1022 | 2 | 1008 | 3 | 2
Guanaco-33B | 1009 | 4 | 992 | 3 | 1002 | 4 | 4
ChatGPT-3.5 Turbo | 916 | 7 | 966 | 5 | 1015 | 2 | 5
Vicuna-13B | 984 | 5 | 974 | 4 | 936 | 5 | 5
Guanaco-13B | 975 | 6 | 913 | 6 | 885 | 6 | 6
Guanaco-7B | 1010 | 3 | 879 | 8 | 860 | 7 | 7
Bard | 909 | 8 | 902 | 7 | - | - | 8
We also use the MMLU benchmark to measure performance on a range of language understanding tasks. This is a multiple-choice benchmark covering 57 tasks including elementary mathematics, US history, computer science, law, and more. We report 5-shot test accuracy.
Dataset | 7B | 13B | 33B | 65B
---|---|---|---|---
LLaMA no tuning | 35.1 | 46.9 | 57.8 | 63.4
Self-Instruct | 36.4 | 33.3 | 53.0 | 56.7
Longform | 32.1 | 43.2 | 56.6 | 59.7
Chip2 | 34.5 | 41.6 | 53.6 | 59.8
HH-RLHF | 34.9 | 44.6 | 55.8 | 60.1
Unnatural Instruct | 41.9 | 48.1 | 57.3 | 61.3
OASST1 (Guanaco) | 36.6 | 46.4 | 57.0 | 62.2
Alpaca | 38.8 | 47.8 | 57.3 | 62.5
FLAN v2 | 44.5 | 51.4 | 59.2 | 63.9
## Risks and Biases
The model can produce factually incorrect output, and should not be relied on to produce factually accurate information. The model was trained on various public datasets; it is possible that this model could generate lewd, biased, or otherwise offensive outputs.
However, we note that finetuning on OASST1 seems to reduce biases as measured on the CrowS dataset. We report here the performance of Guanaco-65B compared to other baseline models on the CrowS dataset.
| | LLaMA-65B | GPT-3 | OPT-175B | Guanaco-65B |
|----------------------|-----------|-------|----------|---------------|
| Gender | 70.6 | 62.6 | 65.7 | **47.5** |
| Religion | {79.0} | 73.3 | 68.6 | **38.7** |
| Race/Color | 57.0 | 64.7 | 68.6 | **45.3** |
| Sexual orientation | {81.0} | 76.2 | 78.6 | **59.1** |
| Age | 70.1 | 64.4 | 67.8 | **36.3** |
| Nationality | 64.2 | 61.6 | 62.9 | **32.4** |
| Disability | 66.7 | 76.7 | 76.7 | **33.9** |
| Physical appearance | 77.8 | 74.6 | 76.2 | **43.1** |
| Socioeconomic status | 71.5 | 73.8 | 76.2 | **55.3** |
| Average | 66.6 | 67.2 | 69.5 | **43.5** |
## Citation
```bibtex
@article{dettmers2023qlora,
title={QLoRA: Efficient Finetuning of Quantized LLMs},
author={Dettmers, Tim and Pagnoni, Artidoro and Holtzman, Ari and Zettlemoyer, Luke},
journal={arXiv preprint arXiv:2305.14314},
year={2023}
}
```
|
ripred/ppo-LunarLander-v2
|
ripred
| 2023-06-13T01:06:55Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-13T01:06:31Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 237.53 +/- 16.48
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
GoGD/xlm-roberta-base-finetuned-panx-en
|
GoGD
| 2023-06-13T01:04:45Z | 125 | 0 |
transformers
|
[
"transformers",
"pytorch",
"xlm-roberta",
"token-classification",
"generated_from_trainer",
"dataset:xtreme",
"license:mit",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
token-classification
| 2023-06-13T00:56:17Z |
---
license: mit
tags:
- generated_from_trainer
datasets:
- xtreme
metrics:
- f1
model-index:
- name: xlm-roberta-base-finetuned-panx-en
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: xtreme
type: xtreme
args: PAN-X.en
metrics:
- name: F1
type: f1
value: 0.7061452513966481
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# xlm-roberta-base-finetuned-panx-en
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the xtreme dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3971
- F1: 0.7061
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 24
- eval_batch_size: 24
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| 1.1723 | 1.0 | 50 | 0.5785 | 0.4900 |
| 0.5347 | 2.0 | 100 | 0.4509 | 0.6298 |
| 0.3791 | 3.0 | 150 | 0.3971 | 0.7061 |
### Framework versions
- Transformers 4.16.2
- Pytorch 1.12.1
- Datasets 1.16.1
- Tokenizers 0.13.3
|
tomoohive/q-taxi-v3
|
tomoohive
| 2023-06-13T00:55:27Z | 0 | 0 | null |
[
"Taxi-v3",
"q-learning",
"reinforcement-learning",
"custom-implementation",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-13T00:55:26Z |
---
tags:
- Taxi-v3
- q-learning
- reinforcement-learning
- custom-implementation
model-index:
- name: q-taxi-v3
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Taxi-v3
type: Taxi-v3
metrics:
- type: mean_reward
value: 7.52 +/- 2.74
name: mean_reward
verified: false
---
# **Q-Learning** Agent playing1 **Taxi-v3**
This is a trained model of a **Q-Learning** agent playing **Taxi-v3** .
## Usage
```python
model = load_from_hub(repo_id="tomoohive/q-taxi-v3", filename="q-learning.pkl")
# Don't forget to check if you need to add additional attributes (is_slippery=False etc)
env = gym.make(model["env_id"])
```
|
tomoohive/q-FrozenLake-v1-4x4-noSlippery
|
tomoohive
| 2023-06-13T00:53:14Z | 0 | 0 | null |
[
"FrozenLake-v1-4x4-no_slippery",
"q-learning",
"reinforcement-learning",
"custom-implementation",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-13T00:53:12Z |
---
tags:
- FrozenLake-v1-4x4-no_slippery
- q-learning
- reinforcement-learning
- custom-implementation
model-index:
- name: q-FrozenLake-v1-4x4-noSlippery
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: FrozenLake-v1-4x4-no_slippery
type: FrozenLake-v1-4x4-no_slippery
metrics:
- type: mean_reward
value: 1.00 +/- 0.00
name: mean_reward
verified: false
---
# **Q-Learning** Agent playing1 **FrozenLake-v1**
This is a trained model of a **Q-Learning** agent playing **FrozenLake-v1** .
## Usage
```python
model = load_from_hub(repo_id="tomoohive/q-FrozenLake-v1-4x4-noSlippery", filename="q-learning.pkl")
# Don't forget to check if you need to add additional attributes (is_slippery=False etc)
env = gym.make(model["env_id"])
```
|
coyude/Chinese-plus-Pygmalion-7b-GPTQ
|
coyude
| 2023-06-13T00:52:09Z | 25 | 29 |
transformers
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"zh",
"en",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T12:37:07Z |
---
license: apache-2.0
language:
- zh
- en
library_name: transformers
pipeline_tag: text-generation
---
原始模型：https://huggingface.co/Neko-Institute-of-Science/pygmalion-7b
lora：https://huggingface.co/ziqingyang/chinese-llama-plus-lora-7b
https://huggingface.co/ziqingyang/chinese-alpaca-plus-lora-7b
将pygmalion-7b与chinese-llama-plus-lora-7b和chinese-alpaca-plus-lora-7b进行合并，增强模型的中文能力，~~不过存在翻译腔~~
使用项目：
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**兼容AutoGPTQ和GPTQ-for-LLaMa**
**若选择GPTQ-for-LLaMa加载，请设置 Wbits=4 groupsize=128 model_type=llama**
Text-generation-webui懒人包：
https://www.bilibili.com/read/cv23495183
---
Original model: https://huggingface.co/Neko-Institute-of-Science/pygmalion-7b
lora：https://huggingface.co/ziqingyang/chinese-llama-plus-lora-7b
https://huggingface.co/ziqingyang/chinese-alpaca-plus-lora-7b
The pygmalion-7b model is combined with the chinese-llama-plus-lora-7b and chinese-alpaca-plus-lora-7b to enhance the model's Chinese language capabilities, ~~although there may be some translated tone~~.
Usage projects:
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If you choose to load GPTQ-for-LLaMa, please set Wbits=4 groupsize=128 model_type=llama**
|
coyude/Chinese-plus-Pygmalion-13b-GPTQ
|
coyude
| 2023-06-13T00:51:22Z | 9 | 10 |
transformers
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"zh",
"en",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T15:21:43Z |
---
license: apache-2.0
language:
- zh
- en
library_name: transformers
pipeline_tag: text-generation
---
原始模型：https://huggingface.co/TehVenom/Pygmalion-13b-Merged
lora：https://huggingface.co/ziqingyang/chinese-llama-plus-lora-13b
https://huggingface.co/ziqingyang/chinese-alpaca-plus-lora-13b
将Pygmalion-13b与chinese-llama-plus-lora-13b和chinese-alpaca-plus-lora-13b进行合并，增强模型的中文能力，~~不过存在翻译腔~~
使用项目：
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**兼容AutoGPTQ和GPTQ-for-LLaMa**
**若选择GPTQ-for-LLaMa加载，请设置 Wbits=4 groupsize=128 model_type=llama**
Text-generation-webui懒人包：
https://www.bilibili.com/read/cv23495183
---
Original model: https://huggingface.co/TehVenom/Pygmalion-13b-Merged
lora：https://huggingface.co/ziqingyang/chinese-llama-plus-lora-13b
https://huggingface.co/ziqingyang/chinese-alpaca-plus-lora-13b
The Pygmalion-13b model is combined with the chinese-alpaca-plus-lora-13b model and chinese-llama-plus-lora-13b model to enhance the model's Chinese language capabilities, ~~although there may be some translated tone~~.
Usage projects:
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If you choose to load GPTQ-for-LLaMa, please set Wbits=4 groupsize=128 model_type=llama**
|
coyude/Chinese-Pygmalion-13b-GPTQ
|
coyude
| 2023-06-13T00:50:08Z | 9 | 2 |
transformers
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"zh",
"en",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T13:21:59Z |
---
license: apache-2.0
language:
- zh
- en
library_name: transformers
pipeline_tag: text-generation
---
原始模型：https://huggingface.co/TehVenom/Pygmalion-13b-Merged
lora：https://huggingface.co/ziqingyang/chinese-alpaca-lora-13b
将Pygmalion-13b与chinese-alpaca-lora-13b进行合并，增强模型的中文能力，~~不过存在翻译腔~~
使用项目：
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**兼容AutoGPTQ和GPTQ-for-LLaMa**
**若选择GPTQ-for-LLaMa加载，请设置 Wbits=4 groupsize=128 model_type=llama**
Text-generation-webui懒人包：
https://www.bilibili.com/read/cv23495183
---
Original model: https://huggingface.co/TehVenom/Pygmalion-13b-Merged
lora：https://huggingface.co/ziqingyang/chinese-alpaca-lora-13b
The Pygmalion-13b model is combined with the chinese-alpaca-lora-13b model to enhance the model's Chinese language capabilities, ~~although there may be some translated tone~~.
Usage projects:
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If you choose to load GPTQ-for-LLaMa, please set Wbits=4 groupsize=128 model_type=llama**
|
coyude/Nous-Hermes-13b-Chinese-plus-GPTQ
|
coyude
| 2023-06-13T00:49:36Z | 16 | 9 |
transformers
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"zh",
"en",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T08:57:07Z |
---
license: apache-2.0
language:
- zh
- en
library_name: transformers
pipeline_tag: text-generation
---
原始模型：https://huggingface.co/NousResearch/Nous-Hermes-13b
lora：https://huggingface.co/ziqingyang/chinese-llama-plus-lora-13b
https://huggingface.co/ziqingyang/chinese-alpaca-plus-lora-13b
将Nous-Hermes-13b与chinese-llama-plus-lora-13b和chinese-alpaca-plus-lora-13b进行合并，增强模型的中文能力，~~不过存在翻译腔~~
使用项目：
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**兼容AutoGPTQ和GPTQ-for-LLaMa**
**若选择GPTQ-for-LLaMa加载，请设置 Wbits=4 groupsize=128 model_type=llama**
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If selecting GPTQ-for-LLaMa loading, please set Wbits=4 groupsize=128 model_type=llama**
Text-generation-webui懒人包：
https://www.bilibili.com/read/cv23495183
---
Original model: https://huggingface.co/NousResearch/Nous-Hermes-13b
lora：https://huggingface.co/ziqingyang/chinese-llama-plus-lora-13b
https://huggingface.co/ziqingyang/chinese-alpaca-plus-lora-13b
The Nous-Hermes-13b model is combined with the chinese-llama-plus-lora-13b and chinese-alpaca-plus-lora-13b to enhance the model's Chinese language capabilities, ~~although there may be some translated tone~~.
Usage projects:
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If you choose to load GPTQ-for-LLaMa, please set Wbits=4 groupsize=128 model_type=llama**
|
coyude/Nous-Hermes-13b-Chinese-GPTQ
|
coyude
| 2023-06-13T00:47:18Z | 11 | 4 |
transformers
|
[
"transformers",
"pytorch",
"llama",
"text-generation",
"zh",
"en",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-11T10:34:33Z |
---
license: apache-2.0
language:
- zh
- en
library_name: transformers
pipeline_tag: text-generation
---
原始模型：https://huggingface.co/NousResearch/Nous-Hermes-13b
lora：https://huggingface.co/ziqingyang/chinese-alpaca-lora-13b
将Nous-Hermes-13b与chinese-alpaca-lora-13b进行合并，增强模型的中文能力，~~不过存在翻译腔~~
使用项目：
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**兼容AutoGPTQ和GPTQ-for-LLaMa**
**若选择GPTQ-for-LLaMa加载，请设置 Wbits=4 groupsize=128 model_type=llama**
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If selecting GPTQ-for-LLaMa loading, please set Wbits=4 groupsize=128 model_type=llama**
Text-generation-webui懒人包：
https://www.bilibili.com/read/cv23495183
---
Original model: https://huggingface.co/NousResearch/Nous-Hermes-13b
Lora: https://huggingface.co/ziqingyang/chinese-alpaca-lora-13b
The Nous-Hermes-13b model is combined with the chinese-alpaca-lora-13b model to enhance the model's Chinese language capabilities, ~~although there may be some translated tone~~.
Usage projects:
https://github.com/ymcui/Chinese-LLaMA-Alpaca
https://github.com/qwopqwop200/GPTQ-for-LLaMa
**Compatible with AutoGPTQ and GPTQ-for-LLaMa**
**If you choose to load GPTQ-for-LLaMa, please set Wbits=4 groupsize=128 model_type=llama**
|
Gilung666/jacklynneubeur
|
Gilung666
| 2023-06-13T00:43:05Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-13T00:42:19Z |
---
license: creativeml-openrail-m
---
|
Rongpi007/Lelua
|
Rongpi007
| 2023-06-13T00:37:34Z | 0 | 0 |
adapter-transformers
|
[
"adapter-transformers",
"chemistry",
"biology",
"medical",
"en",
"dataset:fka/awesome-chatgpt-prompts",
"dataset:tiiuae/falcon-refinedweb",
"arxiv:1910.09700",
"license:bigcode-openrail-m",
"region:us"
] | null | 2023-06-13T00:32:32Z |
---
license: bigcode-openrail-m
datasets:
- fka/awesome-chatgpt-prompts
- tiiuae/falcon-refinedweb
language:
- en
metrics:
- accuracy
- character
library_name: adapter-transformers
tags:
- chemistry
- biology
- medical
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
- **Developed by:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Data Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
gameofdimension/Reinforce-Pixelcopter-PLE-v0
|
gameofdimension
| 2023-06-13T00:30:16Z | 0 | 0 | null |
[
"Pixelcopter-PLE-v0",
"reinforce",
"reinforcement-learning",
"custom-implementation",
"deep-rl-class",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-11T19:16:02Z |
---
tags:
- Pixelcopter-PLE-v0
- reinforce
- reinforcement-learning
- custom-implementation
- deep-rl-class
model-index:
- name: Reinforce-Pixelcopter-PLE-v0
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Pixelcopter-PLE-v0
type: Pixelcopter-PLE-v0
metrics:
- type: mean_reward
value: 99.60 +/- 71.69
name: mean_reward
verified: false
---
# **Reinforce** Agent playing **Pixelcopter-PLE-v0**
This is a trained model of a **Reinforce** agent playing **Pixelcopter-PLE-v0** .
To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction
|
DunnBC22/vit-base-patch16-224-in21k-Brain_Tumors_Image_Classification
|
DunnBC22
| 2023-06-13T00:00:56Z | 191 | 1 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"vit",
"image-classification",
"generated_from_trainer",
"en",
"dataset:imagefolder",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
image-classification
| 2023-06-09T17:28:50Z |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
- f1
- recall
- precision
model-index:
- name: vit-base-patch16-224-in21k-Brain_Tumors_Image_Classification
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.8197969543147208
language:
- en
pipeline_tag: image-classification
---
<h1>vit-base-patch16-224-in21k-Brain_Tumors_Image_Classification</h1>
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k).
It achieves the following results on the evaluation set:
- Loss: 0.8584
- Accuracy: 0.8198
- Weighted f1: 0.7987
- Micro f1: 0.8198
- Macro f1: 0.8054
- Weighted recall: 0.8198
- Micro recall: 0.8198
- Macro recall: 0.8149
- Weighted precision: 0.8615
- Micro precision: 0.8198
- Macro precision: 0.8769
<div style="text-align: center;">
<h2>
Model Description
</h2>
<a href=“https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Brain%20Tumors%20Image%20Classification%20Comparison/Vit%20-%20Image%20Classification.ipynb”>
Click here for the code that I used to create this model.
</a>
This project is part of a comparison of seventeen (17) transformers.
<a href="https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Brain%20Tumors%20Image%20Classification%20Comparison/README.md">
Click here to see the README markdown file for the full project.
</a>
<h2>
Intended Uses & Limitations
</h2>
This model is intended to demonstrate my ability to solve a complex problem using technology.
<br />
<h2>
Training & Evaluation Data
</h2>
<a href="https://www.kaggle.com/datasets/sartajbhuvaji/brain-tumor-classification-mri">
Brain Tumor Image Classification Dataset
</a>
<h2>
Sample Images
</h2>
<img src="https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/raw/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Brain%20Tumors%20Image%20Classification%20Comparison/Images/Sample%20Images.png" />
<h2>
Class Distribution of Training Dataset
</h2>
<img src="https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/raw/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Brain%20Tumors%20Image%20Classification%20Comparison/Images/Class%20Distribution%20-%20Training%20Dataset.png"/>
<h2>
Class Distribution of Evaluation Dataset
</h2>
<img src="https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/raw/main/Computer%20Vision/Image%20Classification/Multiclass%20Classification/Brain%20Tumors%20Image%20Classification%20Comparison/Images/Class%20Distribution%20-%20Testing%20Dataset.png"/>
</div>
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Weighted f1 | Micro f1 | Macro f1 | Weighted recall | Micro recall | Macro recall | Weighted precision | Micro precision | Macro precision |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:-----------:|:--------:|:--------:|:---------------:|:------------:|:------------:|:------------------:|:---------------:|:---------------:|
| 1.3668 | 1.0 | 180 | 1.0736 | 0.6853 | 0.6524 | 0.6853 | 0.6428 | 0.6853 | 0.6853 | 0.6530 | 0.7637 | 0.6853 | 0.7866 |
| 1.3668 | 2.0 | 360 | 1.0249 | 0.7792 | 0.7335 | 0.7792 | 0.7411 | 0.7792 | 0.7792 | 0.7758 | 0.8391 | 0.7792 | 0.8528 |
| 0.1864 | 3.0 | 540 | 0.8584 | 0.8198 | 0.7987 | 0.8198 | 0.8054 | 0.8198 | 0.8198 | 0.8149 | 0.8615 | 0.8198 | 0.8769 |
### Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0
- Datasets 2.11.0
- Tokenizers 0.13.3
|
Josias-Ounsinli/my_awesome_model3
|
Josias-Ounsinli
| 2023-06-12T23:56:43Z | 61 | 0 |
transformers
|
[
"transformers",
"tf",
"distilbert",
"text-classification",
"generated_from_keras_callback",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2023-06-12T23:28:44Z |
---
license: apache-2.0
tags:
- generated_from_keras_callback
model-index:
- name: Josias-Ounsinli/my_awesome_model3
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# Josias-Ounsinli/my_awesome_model3
This model is a fine-tuned version of [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 0.4726
- Validation Loss: 0.6257
- Train Accuracy: 0.7363
- Epoch: 2
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 5625, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
- training_precision: float32
### Training results
| Train Loss | Validation Loss | Train Accuracy | Epoch |
|:----------:|:---------------:|:--------------:|:-----:|
| 0.7231 | 0.6212 | 0.725 | 0 |
| 0.5738 | 0.6079 | 0.727 | 1 |
| 0.4726 | 0.6257 | 0.7363 | 2 |
### Framework versions
- Transformers 4.30.1
- TensorFlow 2.12.0
- Datasets 2.12.0
- Tokenizers 0.13.3
|
gogamza/kobart-base-v2
|
gogamza
| 2023-06-12T23:54:20Z | 7,208 | 32 |
transformers
|
[
"transformers",
"pytorch",
"safetensors",
"bart",
"feature-extraction",
"ko",
"arxiv:1910.13461",
"arxiv:1910.09700",
"license:mit",
"endpoints_compatible",
"region:us"
] |
feature-extraction
| 2022-03-02T23:29:05Z |
---
language: ko
license: mit
tags:
- bart
---
# Model Card for kobart-base-v2
# Model Details
## Model Description
[**BART**](https://arxiv.org/pdf/1910.13461.pdf)(**B**idirectional and **A**uto-**R**egressive **T**ransformers)는 입력 텍스트 일부에 노이즈를 추가하여 이를 다시 원문으로 복구하는 `autoencoder`의 형태로 학습이 됩니다. 한국어 BART(이하 **KoBART**) 는 논문에서 사용된 `Text Infilling` 노이즈 함수를 사용하여 **40GB** 이상의 한국어 텍스트에 대해서 학습한 한국어 `encoder-decoder` 언어 모델입니다. 이를 통해 도출된 `KoBART-base`를 배포합니다.
- **Developed by:** More information needed
- **Shared by [Optional]:** Heewon(Haven) Jeon
- **Model type:** Feature Extraction
- **Language(s) (NLP):** Korean
- **License:** MIT
- **Parent Model:** BART
- **Resources for more information:**
- [GitHub Repo](https://github.com/haven-jeon/KoBART)
- [Model Demo Space](https://huggingface.co/spaces/gogamza/kobart-summarization)
# Uses
## Direct Use
This model can be used for the task of Feature Extraction.
## Downstream Use [Optional]
More information needed.
## Out-of-Scope Use
The model should not be used to intentionally create hostile or alienating environments for people.
# Bias, Risks, and Limitations
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
# Training Details
## Training Data
| Data | # of Sentences |
|-------|---------------:|
| Korean Wiki | 5M |
| Other corpus | 0.27B |
한국어 위키 백과 이외, 뉴스, 책, [모두의 말뭉치 v1.0(대화, 뉴스, ...)](https://corpus.korean.go.kr/), [청와대 국민청원](https://github.com/akngs/petitions) 등의 다양한 데이터가 모델 학습에 사용되었습니다.
`vocab` 사이즈는 30,000 이며 대화에 자주 쓰이는 아래와 같은 이모티콘, 이모지 등을 추가하여 해당 토큰의 인식 능력을 올렸습니다.
> 😀, 😁, 😆, 😅, 🤣, .. , `:-)`, `:)`, `-)`, `(-:`...
## Training Procedure
### Tokenizer
[`tokenizers`](https://github.com/huggingface/tokenizers) 패키지의 `Character BPE tokenizer`로 학습되었습니다.
### Speeds, Sizes, Times
| Model | # of params | Type | # of layers | # of heads | ffn_dim | hidden_dims |
|--------------|:----:|:-------:|--------:|--------:|--------:|--------------:|
| `KoBART-base` | 124M | Encoder | 6 | 16 | 3072 | 768 |
| | | Decoder | 6 | 16 | 3072 | 768 |
# Evaluation
## Testing Data, Factors & Metrics
### Testing Data
More information needed
### Factors
More information needed
### Metrics
More information needed
## Results
NSMC
- acc. : 0.901
The model authors also note in the [GitHub Repo](https://github.com/haven-jeon/KoBART):
| | [NSMC](https://github.com/e9t/nsmc)(acc) | [KorSTS](https://github.com/kakaobrain/KorNLUDatasets)(spearman) | [Question Pair](https://github.com/aisolab/nlp_classification/tree/master/BERT_pairwise_text_classification/qpair)(acc) |
|---|---|---|---|
| **KoBART-base** | 90.24 | 81.66 | 94.34 |
# Model Examination
More information needed
# Environmental Impact
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** More information needed
- **Hours used:** More information needed
- **Cloud Provider:** More information needed
- **Compute Region:** More information needed
- **Carbon Emitted:** More information needed
# Technical Specifications [optional]
## Model Architecture and Objective
More information needed
## Compute Infrastructure
More information needed
### Hardware
More information needed
### Software
More information needed.
# Citation
**BibTeX:**
More information needed.
# Glossary [optional]
More information needed
# More Information [optional]
More information needed
# Model Card Authors [optional]
Heewon(Haven) Jeon in collaboration with Ezi Ozoani and the Hugging Face team
# Model Card Contact
The model authors note in the [GitHub Repo](https://github.com/haven-jeon/KoBART):
`KoBART` 관련 이슈는 [이곳](https://github.com/SKT-AI/KoBART/issues)에 올려주세요.
# How to Get Started with the Model
Use the code below to get started with the model.
<details>
<summary> Click to expand </summary>
```python
from transformers import PreTrainedTokenizerFast, BartModel
tokenizer = PreTrainedTokenizerFast.from_pretrained('gogamza/kobart-base-v2')
model = BartModel.from_pretrained('gogamza/kobart-base-v2')
```
</details>
|
gogamza/kobart-summarization
|
gogamza
| 2023-06-12T23:54:06Z | 4,850 | 11 |
transformers
|
[
"transformers",
"pytorch",
"safetensors",
"bart",
"text2text-generation",
"ko",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2022-03-02T23:29:05Z |
---
language: ko
tags:
- bart
license: mit
---
# Korean News Summarization Model
## Demo
https://huggingface.co/spaces/gogamza/kobart-summarization
## How to use
```python
import torch
from transformers import PreTrainedTokenizerFast
from transformers import BartForConditionalGeneration
tokenizer = PreTrainedTokenizerFast.from_pretrained('gogamza/kobart-summarization')
model = BartForConditionalGeneration.from_pretrained('gogamza/kobart-summarization')
text = "과거를 떠올려보자. 방송을 보던 우리의 모습을. 독보적인 매체는 TV였다. 온 가족이 둘러앉아 TV를 봤다. 간혹 가족들끼리 뉴스와 드라마, 예능 프로그램을 둘러싸고 리모컨 쟁탈전이 벌어지기도 했다. 각자 선호하는 프로그램을 ‘본방’으로 보기 위한 싸움이었다. TV가 한 대인지 두 대인지 여부도 그래서 중요했다. 지금은 어떤가. ‘안방극장’이라는 말은 옛말이 됐다. TV가 없는 집도 많다. 미디어의 혜 택을 누릴 수 있는 방법은 늘어났다. 각자의 방에서 각자의 휴대폰으로, 노트북으로, 태블릿으로 콘텐츠 를 즐긴다."
raw_input_ids = tokenizer.encode(text)
input_ids = [tokenizer.bos_token_id] + raw_input_ids + [tokenizer.eos_token_id]
summary_ids = model.generate(torch.tensor([input_ids]))
tokenizer.decode(summary_ids.squeeze().tolist(), skip_special_tokens=True)
```
|
renyulin/dqn-SpaceInvadersNoFrameskip-v4
|
renyulin
| 2023-06-12T23:49:58Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"SpaceInvadersNoFrameskip-v4",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T23:49:28Z |
---
library_name: stable-baselines3
tags:
- SpaceInvadersNoFrameskip-v4
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: DQN
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: SpaceInvadersNoFrameskip-v4
type: SpaceInvadersNoFrameskip-v4
metrics:
- type: mean_reward
value: 29.00 +/- 64.30
name: mean_reward
verified: false
---
# **DQN** Agent playing **SpaceInvadersNoFrameskip-v4**
This is a trained model of a **DQN** agent playing **SpaceInvadersNoFrameskip-v4**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
The RL Zoo is a training framework for Stable Baselines3
reinforcement learning agents,
with hyperparameter optimization and pre-trained agents included.
## Usage (with SB3 RL Zoo)
RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo<br/>
SB3: https://github.com/DLR-RM/stable-baselines3<br/>
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
Install the RL Zoo (with SB3 and SB3-Contrib):
```bash
pip install rl_zoo3
```
```
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga renyulin -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
```
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga renyulin -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
## Training (with the RL Zoo)
```
python -m rl_zoo3.train --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga renyulin
```
## Hyperparameters
```python
OrderedDict([('batch_size', 32),
('buffer_size', 100000),
('env_wrapper',
['stable_baselines3.common.atari_wrappers.AtariWrapper']),
('exploration_final_eps', 0.01),
('exploration_fraction', 0.1),
('frame_stack', 4),
('gradient_steps', 1),
('learning_rate', 0.0001),
('learning_starts', 100000),
('n_timesteps', 100000.0),
('optimize_memory_usage', False),
('policy', 'CnnPolicy'),
('target_update_interval', 1000),
('train_freq', 4),
('normalize', False)])
```
# Environment Arguments
```python
{'render_mode': 'rgb_array'}
```
|
walkerrose/cv_summarization-distilbart-cnn-16-6
|
walkerrose
| 2023-06-12T22:47:22Z | 74 | 0 |
transformers
|
[
"transformers",
"tf",
"bart",
"text2text-generation",
"generated_from_keras_callback",
"summarization",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
summarization
| 2023-06-12T10:18:44Z |
---
license: apache-2.0
tags:
- generated_from_keras_callback
- summarization
model-index:
- name: cv_summarization-distilbart-cnn-16-6
results: []
language:
- en
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# cv_summarization-distilbart-cnn-16-6
This model is a fine-tuned version of [philschmid/tf-distilbart-cnn-12-6](https://huggingface.co/philschmid/tf-distilbart-cnn-12-6) on [cnn_dailymail](https://huggingface.co/datasets/cnn_dailymail) dataset.
It achieves the following results on the evaluation set:
- Train Loss: 0.1445
- Validation Loss: 0.4524
- Epoch: 4
- Rouge1: 76.90
- Rouge2: 73.30
- RougeL: 76.04
- RougeLsum: 75.87
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': 2e-05}
- training_precision: float32
### Training results
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 0.7265 | 0.4665 | 0 |
| 0.3192 | 0.4480 | 1 |
| 0.2154 | 0.4395 | 2 |
| 0.1587 | 0.4668 | 3 |
| 0.1445 | 0.4524 | 4 |
### Framework versions
- Transformers 4.30.1
- TensorFlow 2.12.0
- Datasets 2.12.0
- Tokenizers 0.13.3
|
Uminosachi/sam-hq
|
Uminosachi
| 2023-06-12T22:13:19Z | 0 | 0 | null |
[
"arxiv:2306.01567",
"license:apache-2.0",
"region:us"
] | null | 2023-06-12T21:22:36Z |
---
license: apache-2.0
---
Three models of [Segment Anything in High Quality](https://github.com/SysCV/sam-hq).
**Reference**
Ke, L., Ye, M., Danelljan, M., Liu, Y., Tai, Y-W., Tang, C-K., & Yu, F. (2023). [Segment Anything in High Quality](https://arxiv.org/abs/2306.01567). arXiv:2306.01567.
|
pcapp/dqn-SpaceInvadersNoFrameskip-v4
|
pcapp
| 2023-06-12T21:45:29Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"SpaceInvadersNoFrameskip-v4",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T21:43:25Z |
---
library_name: stable-baselines3
tags:
- SpaceInvadersNoFrameskip-v4
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: DQN
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: SpaceInvadersNoFrameskip-v4
type: SpaceInvadersNoFrameskip-v4
metrics:
- type: mean_reward
value: 587.00 +/- 265.68
name: mean_reward
verified: false
---
# **DQN** Agent playing **SpaceInvadersNoFrameskip-v4**
This is a trained model of a **DQN** agent playing **SpaceInvadersNoFrameskip-v4**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
The RL Zoo is a training framework for Stable Baselines3
reinforcement learning agents,
with hyperparameter optimization and pre-trained agents included.
## Usage (with SB3 RL Zoo)
RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo<br/>
SB3: https://github.com/DLR-RM/stable-baselines3<br/>
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
Install the RL Zoo (with SB3 and SB3-Contrib):
```bash
pip install rl_zoo3
```
```
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga pcapp -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
```
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga pcapp -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
## Training (with the RL Zoo)
```
python -m rl_zoo3.train --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga pcapp
```
## Hyperparameters
```python
OrderedDict([('batch_size', 32),
('buffer_size', 100000),
('env_wrapper',
['stable_baselines3.common.atari_wrappers.AtariWrapper']),
('exploration_final_eps', 0.01),
('exploration_fraction', 0.1),
('frame_stack', 4),
('gradient_steps', 1),
('learning_rate', 0.0001),
('learning_starts', 100000),
('n_timesteps', 1000000.0),
('optimize_memory_usage', False),
('policy', 'CnnPolicy'),
('target_update_interval', 1000),
('train_freq', 4),
('normalize', False)])
```
# Environment Arguments
```python
{'render_mode': 'rgb_array'}
```
|
gokuls/bert_base_48
|
gokuls
| 2023-06-12T21:44:52Z | 140 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"bert",
"fill-mask",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
fill-mask
| 2023-06-11T21:10:41Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: bert_base_48
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert_base_48
This model is a fine-tuned version of [gokuls/bert_base_24](https://huggingface.co/gokuls/bert_base_24) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 5.6418
- Accuracy: 0.1663
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 48
- eval_batch_size: 48
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 6.005 | 0.08 | 10000 | 5.9999 | 0.1513 |
| 5.9845 | 0.16 | 20000 | 5.9811 | 0.1511 |
| 5.9695 | 0.25 | 30000 | 5.9673 | 0.1518 |
| 5.9599 | 0.33 | 40000 | 5.9513 | 0.1515 |
| 5.9402 | 0.41 | 50000 | 5.9411 | 0.1517 |
| 5.9291 | 0.49 | 60000 | 5.9217 | 0.1518 |
| 5.6816 | 0.57 | 70000 | 5.6418 | 0.1663 |
### Framework versions
- Transformers 4.30.1
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3
|
mihirdeo16/poca-SoccerTwos
|
mihirdeo16
| 2023-06-12T21:28:13Z | 16 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"SoccerTwos",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-SoccerTwos",
"region:us"
] |
reinforcement-learning
| 2023-06-12T21:28:03Z |
---
library_name: ml-agents
tags:
- SoccerTwos
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-SoccerTwos
---
# **poca** Agent playing **SoccerTwos**
This is a trained model of a **poca** agent playing **SoccerTwos**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: mihirdeo16/poca-SoccerTwos
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
Maxlumaga/taniav2
|
Maxlumaga
| 2023-06-12T21:19:51Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-11T15:06:54Z |
---
license: creativeml-openrail-m
---
|
fgeyer/dqn-SpaceInvadersNoFrameskip-v4
|
fgeyer
| 2023-06-12T21:03:52Z | 2 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"SpaceInvadersNoFrameskip-v4",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T21:02:41Z |
---
library_name: stable-baselines3
tags:
- SpaceInvadersNoFrameskip-v4
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: DQN
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: SpaceInvadersNoFrameskip-v4
type: SpaceInvadersNoFrameskip-v4
metrics:
- type: mean_reward
value: 783.50 +/- 478.73
name: mean_reward
verified: false
---
# **DQN** Agent playing **SpaceInvadersNoFrameskip-v4**
This is a trained model of a **DQN** agent playing **SpaceInvadersNoFrameskip-v4**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
The RL Zoo is a training framework for Stable Baselines3
reinforcement learning agents,
with hyperparameter optimization and pre-trained agents included.
## Usage (with SB3 RL Zoo)
RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo<br/>
SB3: https://github.com/DLR-RM/stable-baselines3<br/>
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
Install the RL Zoo (with SB3 and SB3-Contrib):
```bash
pip install rl_zoo3
```
```
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga fgeyer -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
```
python -m rl_zoo3.load_from_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -orga fgeyer -f logs/
python -m rl_zoo3.enjoy --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
```
## Training (with the RL Zoo)
```
python -m rl_zoo3.train --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f logs/ -orga fgeyer
```
## Hyperparameters
```python
OrderedDict([('batch_size', 32),
('buffer_size', 100000),
('env_wrapper',
['stable_baselines3.common.atari_wrappers.AtariWrapper']),
('exploration_final_eps', 0.01),
('exploration_fraction', 0.1),
('frame_stack', 4),
('gradient_steps', 1),
('learning_rate', 0.0001),
('learning_starts', 100000),
('n_timesteps', 10000000.0),
('optimize_memory_usage', False),
('policy', 'CnnPolicy'),
('target_update_interval', 1000),
('train_freq', 4),
('normalize', False)])
```
# Environment Arguments
```python
{'render_mode': 'rgb_array'}
```
|
guibvieira/topic_modelling
|
guibvieira
| 2023-06-12T21:03:48Z | 9 | 0 |
bertopic
|
[
"bertopic",
"text-classification",
"region:us"
] |
text-classification
| 2023-06-12T18:18:30Z |
---
tags:
- bertopic
library_name: bertopic
pipeline_tag: text-classification
---
# topic_modelling
This is a [BERTopic](https://github.com/MaartenGr/BERTopic) model.
BERTopic is a flexible and modular topic modeling framework that allows for the generation of easily interpretable topics from large datasets.
## Usage
To use this model, please install BERTopic:
```
pip install -U bertopic
```
You can use the model as follows:
```python
from bertopic import BERTopic
topic_model = BERTopic.load("guibvieira/topic_modelling")
topic_model.get_topic_info()
```
## Topic overview
* Number of topics: 6286
* Number of training documents: 709072
<details>
<summary>Click here for an overview of all topics.</summary>
| Topic ID | Topic Keywords | Topic Frequency | Label |
|----------|----------------|-----------------|-------|
| -1 | liquidity - coins - bro - sorry - morning | 15 | -1_liquidity_coins_bro_sorry |
| 0 | token - tokens - supply - value - price | 241723 | 0_token_tokens_supply_value |
| 1 | main - problem - bro - tho - brother | 3993 | 1_main_problem_bro_tho |
| 2 | twitter - comment - space - social - media | 2185 | 2_twitter_comment_space_social |
| 3 | project - projects - interesting - research - mate | 2122 | 3_project_projects_interesting_research |
| 4 | nft - cool - unique - digital - games | 2100 | 4_nft_cool_unique_digital |
| 5 | coins - coin - shit - sell - value | 2022 | 5_coins_coin_shit_sell |
| 6 | market - situation - bad - control - current | 1999 | 6_market_situation_bad_control |
| 7 | binance - option - account - spot - app | 1819 | 7_binance_option_account_spot |
| 8 | btc - break - short - likely - hit | 1755 | 8_btc_break_short_likely |
| 9 | dm - reply - sir - details - okay | 1735 | 9_dm_reply_sir_details |
| 10 | rewards - reward - event - hour - sir | 1647 | 10_rewards_reward_event_hour |
| 11 | hard - massive - little - huge - opinion | 1638 | 11_hard_massive_little_huge |
| 12 | listing - related - exchange - certain - list | 1634 | 12_listing_related_exchange_certain |
| 13 | wallet - address - site - longer - problem | 1600 | 13_wallet_address_site_longer |
| 14 | investors - investing - research - term - early | 1582 | 14_investors_investing_research_term |
| 15 | talk - use - words - chat - reply | 1535 | 15_talk_use_words_chat |
| 16 | buy - selling - late - lol - yesterday | 1456 | 16_buy_selling_late_lol |
| 17 | crypto - space - world - exciting - projects | 1428 | 17_crypto_space_world_exciting |
| 18 | games - game - play - fun - kind | 1406 | 18_games_game_play_fun |
| 19 | quick - funds - profits - opportunity - active | 1389 | 19_quick_funds_profits_opportunity |
| 20 | issue - problem - solution - fix - problems | 1371 | 20_issue_problem_solution_fix |
| 21 | left - okay - lol - yesterday - man | 1309 | 21_left_okay_lol_yesterday |
| 22 | dear - safe - friends - limited - service | 1293 | 22_dear_safe_friends_limited |
| 23 | exchange - listing - list - major - volume | 1264 | 23_exchange_listing_list_major |
| 24 | short - chart - break - binance - btc | 1212 | 24_short_chart_break_binance |
| 25 | main - fees - profit - market - trade | 1205 | 25_main_fees_profit_market |
| 26 | group - guy - sorry - add - kind | 1184 | 26_group_guy_sorry_add |
| 27 | updates - update - latest - buddy - new | 1137 | 27_updates_update_latest_buddy |
| 28 | second - transactions - address - binance - investment | 1136 | 28_second_transactions_address_binance |
| 29 | withdraw - withdrawal - sir - minutes - issues | 1122 | 29_withdraw_withdrawal_sir_minutes |
| 30 | change - nice - cool - sorry - dude | 1119 | 30_change_nice_cool_sorry |
| 31 | night - early - late - morning - haha | 1104 | 31_night_early_late_morning |
| 32 | trade - trading - spot - short - buddy | 1088 | 32_trade_trading_spot_short |
| 33 | ama - questions - yesterday - upcoming - question | 1056 | 33_ama_questions_yesterday_upcoming |
| 34 | app - version - update - play - web | 1053 | 34_app_version_update_play |
| 35 | sounds - nice - cool - head - awesome | 1041 | 35_sounds_nice_cool_head |
| 36 | ai - projects - interesting - technology - stuff | 1022 | 36_ai_projects_interesting_technology |
| 37 | minutes - normal - yesterday - long - transaction | 1004 | 37_minutes_normal_yesterday_long |
| 38 | announcement - official - mate - channel - moment | 991 | 38_announcement_official_mate_channel |
| 39 | project - guys - buddy - ready - tbh | 975 | 39_project_guys_buddy_ready |
| 40 | admin - answer - reply - question - help | 972 | 40_admin_answer_reply_question |
| 41 | blockchain - technology - data - major - transactions | 959 | 41_blockchain_technology_data_major |
| 42 | ways - unique - small - family - fake | 934 | 42_ways_unique_small_family |
| 43 | price - current - increase - low - drop | 925 | 43_price_current_increase_low |
| 44 | chain - moment - projects - shit - launch | 918 | 44_chain_moment_projects_shit |
| 45 | site - option - period - feature - single | 907 | 45_site_option_period_feature |
| 46 | access - complete - email - times - trust | 899 | 46_access_complete_email_times |
| 47 | fine - dear - nice - country - bro | 897 | 47_fine_dear_nice_country |
| 48 | dev - team - answer - job - sense | 895 | 48_dev_team_answer_job |
| 49 | profit - profits - half - huge - small | 890 | 49_profit_profits_half_huge |
| 50 | number - reply - support - email - sir | 884 | 50_number_reply_support_email |
| 51 | sir - holders - announcement - early - moment | 853 | 51_sir_holders_announcement_early |
| 52 | sense - fun - sorry - guy - dude | 849 | 52_sense_fun_sorry_guy |
| 53 | great - success - low - life - click | 834 | 53_great_success_low_life |
| 54 | selling - biggest - big - example - volume | 833 | 54_selling_biggest_big_example |
| 55 | question - answer - questions - okay - free | 828 | 55_question_answer_questions_okay |
| 56 | country - half - fact - example - sense | 828 | 56_country_half_fact_example |
| 57 | massive - earning - fast - big - positive | 824 | 57_massive_earning_fast_big |
| 58 | bot - reply - damn - admin - shit | 821 | 58_bot_reply_damn_admin |
| 59 | news - bad - good - latest - week | 814 | 59_news_bad_good_latest |
| 60 | small - biggest - couple - tho - man | 799 | 60_small_biggest_couple_tho |
| 61 | period - coins - option - rewards - bit | 794 | 61_period_coins_option_rewards |
| 62 | guy - shit - lol - bro - man | 787 | 62_guy_shit_lol_bro |
| 63 | god - word - man - true - old | 786 | 63_god_word_man_true |
| 64 | period - similar - number - specific - close | 783 | 64_period_similar_number_specific |
| 65 | community - glad - welcome - strong - building | 779 | 65_community_glad_welcome_strong |
| 66 | buying - yesterday - hold - thinking - imo | 755 | 66_buying_yesterday_hold_thinking |
| 67 | listing - list - control - article - volume | 744 | 67_listing_list_control_article |
| 68 | team - hard - work - trust - clear | 732 | 68_team_hard_work_trust |
| 69 | chain - swap - smart - network - worth | 729 | 69_chain_swap_smart_network |
| 70 | message - check - updates - latest - announcement | 722 | 70_message_check_updates_latest |
| 71 | damn - course - clear - research - example | 717 | 71_damn_course_clear_research |
| 72 | eth - chain - short - opinion - hold | 714 | 72_eth_chain_short_opinion |
| 73 | fake - guy - nice - face - sorry | 699 | 73_fake_guy_nice_face |
| 74 | telegram - discord - forward - group - chat | 695 | 74_telegram_discord_forward_group |
| 75 | limited - online - network - thanks - world | 682 | 75_limited_online_network_thanks |
| 76 | block - holders - pool - clear - weeks | 678 | 76_block_holders_pool_clear |
| 77 | claim - option - biggest - earlier - period | 658 | 77_claim_option_biggest_earlier |
| 78 | swap - feature - fee - transaction - fees | 658 | 78_swap_feature_fee_transaction |
| 79 | sir - problem - email - process - account | 652 | 79_sir_problem_email_process |
| 80 | - - - - | 627 | 80____ |
| 81 | pepe - fake - shit - real - run | 623 | 81_pepe_fake_shit_real |
| 82 | contract - address - correct - old - sorry | 617 | 82_contract_address_correct_old |
| 83 | exchange - listing - spot - wallet - sign | 614 | 83_exchange_listing_spot_wallet |
| 84 | gon - shit - fun - end - man | 614 | 84_gon_shit_fun_end |
| 85 | launch - date - fair - product - release | 608 | 85_launch_date_fair_product |
| 86 | word - man - means - words - true | 607 | 86_word_man_means_words |
| 87 | user - change - care - months - dev | 606 | 87_user_change_care_months |
| 88 | phone - data - type - normal - update | 604 | 88_phone_data_type_normal |
| 89 | bot - trading - spot - trade - run | 600 | 89_bot_trading_spot_trade |
| 90 | future - love - amazing - support - simple | 593 | 90_future_love_amazing_support |
| 91 | chain - stuff - main - issues - half | 591 | 91_chain_stuff_main_issues |
| 92 | small - fake - happy - life - big | 578 | 92_small_fake_happy_life |
| 93 | talk - interested - sorry - normal - okay | 574 | 93_talk_interested_sorry_normal |
| 94 | swap - liquidity - contract - address - buy | 566 | 94_swap_liquidity_contract_address |
| 95 | lots - accounts - click - offer - month | 560 | 95_lots_accounts_click_offer |
| 96 | job - benefits - process - extra - blockchain | 557 | 96_job_benefits_process_extra |
| 97 | buying - research - launch - live - mate | 556 | 97_buying_research_launch_live |
| 98 | reward - event - ecosystem - volume - user | 547 | 98_reward_event_ecosystem_volume |
| 99 | - - - - | 544 | 99____ |
| 100 | event - participate - upcoming - buddy - excited | 539 | 100_event_participate_upcoming_buddy |
| 101 | period - months - option - year - feature | 535 | 101_period_months_option_year |
| 102 | announcement - recent - projects - couple - excited | 535 | 102_announcement_recent_projects_couple |
| 103 | projects - building - game - development - features | 534 | 103_projects_building_game_development |
| 104 | deposit - issue - yesterday - admin - issues | 534 | 104_deposit_issue_yesterday_admin |
| 105 | build - dev - phone - bro - use | 513 | 105_build_dev_phone_bro |
| 106 | sure - bro - buddy - brother - man | 509 | 106_sure_bro_buddy_brother |
| 107 | pool - old - rest - live - release | 507 | 107_pool_old_rest_live |
| 108 | mate - guys - yesterday - kinda - fun | 504 | 108_mate_guys_yesterday_kinda |
| 109 | projects - lol - shit - moment - dude | 502 | 109_projects_lol_shit_moment |
| 110 | major - site - minutes - block - half | 502 | 110_major_site_minutes_block |
| 111 | course - year - cool - high - nice | 500 | 111_course_year_cool_high |
| 112 | space - stuff - talk - dude - interesting | 495 | 112_space_stuff_talk_dude |
| 113 | order - position - close - sell - price | 492 | 113_order_position_close_sell |
| 114 | hold - holding - longer - term - long | 488 | 114_hold_holding_longer_term |
| 115 | coins - coin - liquidity - term - hold | 484 | 115_coins_coin_liquidity_term |
| 116 | chart - look - half - damn - shit | 482 | 116_chart_look_half_damn |
| 117 | list - dev - update - moment - form | 480 | 117_list_dev_update_moment |
| 118 | channel - comment - official - step - support | 479 | 118_channel_comment_official_step |
| 119 | reply - sorry - wrong - yesterday - person | 477 | 119_reply_sorry_wrong_yesterday |
| 120 | public - ones - yesterday - open - glad | 476 | 120_public_ones_yesterday_open |
| 121 | friend - friends - dear - dude - new | 473 | 121_friend_friends_dear_dude |
| 122 | morning - good - day - nice - buddy | 473 | 122_morning_good_day_nice |
| 123 | minutes - chance - service - available - telegram | 470 | 123_minutes_chance_service_available |
| 124 | hard - normal - dev - shit - major | 464 | 124_hard_normal_dev_shit |
| 125 | lots - wrong - bad - security - successful | 463 | 125_lots_wrong_bad_security |
| 126 | early - super - low - guys - half | 463 | 126_early_super_low_guys |
| 127 | power - share - trade - investors - investment | 461 | 127_power_share_trade_investors |
| 128 | country - nice - special - welcome - love | 460 | 128_country_nice_special_welcome |
| 129 | morning - break - home - near - fine | 460 | 129_morning_break_home_near |
| 130 | platform - games - features - cool - simple | 458 | 130_platform_games_features_cool |
| 131 | complete - reward - address - pool - details | 457 | 131_complete_reward_address_pool |
| 132 | check - moment - bro - mate - brother | 445 | 132_check_moment_bro_mate |
| 133 | sell - volume - country - longer - rate | 441 | 133_sell_volume_country_longer |
| 134 | plan - plans - future - today - mate | 439 | 134_plan_plans_future_today |
| 135 | market - current - small - low - early | 439 | 135_market_current_small_low |
| 136 | block - sorry - reason - damn - guy | 428 | 136_block_sorry_reason_damn |
| 137 | - - - - | 424 | 137____ |
| 138 | - - - - | 420 | 138____ |
| 139 | things - hard - stuff - way - dude | 420 | 139_things_hard_stuff_way |
| 140 | drop - personal - pm - thanks - time | 418 | 140_drop_personal_pm_thanks |
| 141 | alot - words - security - successful - capital | 415 | 141_alot_words_security_successful |
| 142 | account - accounts - multiple - old - user | 410 | 142_account_accounts_multiple_old |
| 143 | care - potential - marketing - high - real | 409 | 143_care_potential_marketing_high |
| 144 | issue - problem - phone - look - pm | 406 | 144_issue_problem_phone_look |
| 145 | discord - questions - ppl - reply - opinion | 406 | 145_discord_questions_ppl_reply |
| 146 | form - email - business - contact - details | 405 | 146_form_email_business_contact |
| 147 | money - ways - earning - alot - easy | 399 | 147_money_ways_earning_alot |
| 148 | place - tomorrow - active - holding - services | 398 | 148_place_tomorrow_active_holding |
| 149 | man - guy - people - sense - country | 397 | 149_man_guy_people_sense |
| 150 | exciting - upcoming - perfect - ones - word | 395 | 150_exciting_upcoming_perfect_ones |
| 151 | doubt - legit - security - process - easy | 394 | 151_doubt_legit_security_process |
| 152 | claim - pay - swap - version - type | 393 | 152_claim_pay_swap_version |
| 153 | tg - dev - group - cool - dude | 393 | 153_tg_dev_group_cool |
| 154 | info - profits - work - dm - | 393 | 154_info_profits_work_dm |
| 155 | smart - case - post - ecosystem - technology | 392 | 155_smart_case_post_ecosystem |
| 156 | personal - secure - pay - telegram - chat | 391 | 156_personal_secure_pay_telegram |
| 157 | guy - reason - cause - lol - rules | 389 | 157_guy_reason_cause_lol |
| 158 | add - network - clear - problems - fine | 389 | 158_add_network_clear_problems |
| 159 | seeing - different - opportunity - coin - profit | 385 | 159_seeing_different_opportunity_coin |
| 160 | try - thanks - tomorrow - mate - cool | 384 | 160_try_thanks_tomorrow_mate |
| 161 | wait - long - bit - worth - moment | 383 | 161_wait_long_bit_worth |
| 162 | research - important - mate - cool - course | 382 | 162_research_important_mate_cool |
| 163 | love - safe - old - liquidity - contact | 381 | 163_love_safe_old_liquidity |
| 164 | shit - admin - action - message - block | 380 | 164_shit_admin_action_message |
| 165 | pretty - face - person - looks - guy | 373 | 165_pretty_face_person_looks |
| 166 | use - easy - bro - months - excited | 373 | 166_use_easy_bro_months |
| 167 | holding - capital - legit - plan - profits | 372 | 167_holding_capital_legit_plan |
| 168 | form - transaction - safe - payment - fast | 372 | 168_form_transaction_safe_payment |
| 169 | deal - likely - early - situation - global | 370 | 169_deal_likely_early_situation |
| 170 | things - current - updates - update - release | 368 | 170_things_current_updates_update |
| 171 | live - public - buddy - specific - version | 367 | 171_live_public_buddy_specific |
| 172 | question - ama - key - situation - comment | 364 | 172_question_ama_key_situation |
| 173 | idk - sorry - past - stuff - bro | 364 | 173_idk_sorry_past_stuff |
| 174 | accounts - potential - welcome - real - money | 362 | 174_accounts_potential_welcome_real |
| 175 | video - youtube - specific - awesome - short | 361 | 175_video_youtube_specific_awesome |
| 176 | damn - tho - person - night - cool | 361 | 176_damn_tho_person_night |
| 177 | look - thanks - cool - mate - tho | 360 | 177_look_thanks_cool_mate |
| 178 | marketing - product - couple - plan - point | 360 | 178_marketing_product_couple_plan |
| 179 | weeks - old - idea - alot - sense | 359 | 179_weeks_old_idea_alot |
| 180 | lose - huge - situation - year - point | 357 | 180_lose_huge_situation_year |
| 181 | list - cool - idea - kind - bot | 355 | 181_list_cool_idea_kind |
| 182 | form - announcement - option - article - clear | 355 | 182_form_announcement_option_article |
| 183 | early - half - worth - price - drop | 353 | 183_early_half_worth_price |
| 184 | word - words - term - bad - dude | 350 | 184_word_words_term_bad |
| 185 | liquidity - sense - low - situation - hand | 348 | 185_liquidity_sense_low_situation |
| 186 | ready - time - long - bro - cool | 347 | 186_ready_time_long_bro |
| 187 | discord - private - email - ethereum - services | 346 | 187_discord_private_email_ethereum |
| 188 | worth - fake - capital - blockchain - use | 344 | 188_worth_fake_capital_blockchain |
| 189 | lots - legit - true - chat - lot | 344 | 189_lots_legit_true_chat |
| 190 | way - half - anymore - buying - option | 343 | 190_way_half_anymore_buying |
| 191 | assets - situation - problem - funds - shit | 342 | 191_assets_situation_problem_funds |
| 192 | private - chat - talk - message - personal | 342 | 192_private_chat_talk_message |
| 193 | listing - global - volume - exchange - list | 342 | 193_listing_global_volume_exchange |
| 194 | action - information - active - link - | 341 | 194_action_information_active_link |
| 195 | transaction - transactions - clear - normal - issue | 341 | 195_transaction_transactions_clear_normal |
| 196 | situation - country - cool - old - group | 340 | 196_situation_country_cool_old |
| 197 | words - key - option - private - wallet | 339 | 197_words_key_option_private |
| 198 | country - problem - site - phone - issues | 339 | 198_country_problem_site_phone |
| 199 | luck - bad - guy - personal - ones | 339 | 199_luck_bad_guy_personal |
| 200 | fee - legit - fake - stop - extra | 338 | 200_fee_legit_fake_stop |
| 201 | fix - version - video - works - phone | 336 | 201_fix_version_video_works |
| 202 | shit - people - sense - problem - bad | 335 | 202_shit_people_sense_problem |
| 203 | defi - projects - future - problems - benefits | 334 | 203_defi_projects_future_problems |
| 204 | half - rules - chance - specific - lose | 331 | 204_half_rules_chance_specific |
| 205 | fair - country - man - guy - person | 331 | 205_fair_country_man_guy |
| 206 | comment - ama - love - family - holders | 329 | 206_comment_ama_love_family |
| 207 | situation - protocol - issues - feature - funds | 327 | 207_situation_protocol_issues_feature |
| 208 | fee - fees - high - higher - low | 327 | 208_fee_fees_high_higher |
| 209 | information - article - details - website - info | 325 | 209_information_article_details_website |
| 210 | profits - sure - profit - investors - withdrawal | 325 | 210_profits_sure_profit_investors |
| 211 | man - years - guy - old - crazy | 325 | 211_man_years_guy_old |
| 212 | clear - issue - fine - different - page | 323 | 212_clear_issue_fine_different |
| 213 | dev - stuff - community - protocol - idea | 321 | 213_dev_stuff_community_protocol |
| 214 | man - win - real - half - form | 320 | 214_man_win_real_half |
| 215 | app - account - dude - user - problems | 318 | 215_app_account_dude_user |
| 216 | single - true - half - people - guy | 317 | 216_single_true_half_people |
| 217 | supply - total - current - increase - sense | 316 | 217_supply_total_current_increase |
| 218 | bitcoin - technology - interesting - situation - article | 315 | 218_bitcoin_technology_interesting_situation |
| 219 | okay - personal - wan - comment - kinda | 312 | 219_okay_personal_wan_comment |
| 220 | luck - win - comment - event - site | 311 | 220_luck_win_comment_event |
| 221 | assets - control - sense - information - months | 310 | 221_assets_control_sense_information |
| 222 | projects - space - dev - games - news | 309 | 222_projects_space_dev_games |
| 223 | usdt - sell - buying - exchange - btc | 309 | 223_usdt_sell_buying_exchange |
| 224 | technology - opinion - wonder - development - near | 307 | 224_technology_opinion_wonder_development |
| 225 | period - listing - kind - utc - tho | 307 | 225_period_listing_kind_utc |
| 226 | country - dude - half - shit - guy | 306 | 226_country_dude_half_shit |
| 227 | amazing - success - info - support - | 306 | 227_amazing_success_info_support |
| 228 | turn - option - fix - hand - left | 306 | 228_turn_option_fix_hand |
| 229 | dev - shit - word - projects - dude | 305 | 229_dev_shit_word_projects |
| 230 | active - page - big - day - | 303 | 230_active_page_big_day |
| 231 | value - month - worry - service - info | 303 | 231_value_month_worry_service |
| 232 | upcoming - projects - utc - date - period | 303 | 232_upcoming_projects_utc_date |
| 233 | positive - active - page - channel - low | 302 | 233_positive_active_page_channel |
| 234 | capital - update - release - action - business | 300 | 234_capital_update_release_action |
| 235 | message - specific - issue - code - type | 300 | 235_message_specific_issue_code |
| 236 | support - email - reply - contact - team | 299 | 236_support_email_reply_contact |
| 237 | opportunity - missed - kind - situation - guys | 296 | 237_opportunity_missed_kind_situation |
| 238 | true - follow - opportunity - holding - investors | 295 | 238_true_follow_opportunity_holding |
| 239 | ready - care - high - - | 295 | 239_ready_care_high_ |
| 240 | fast - - - - | 295 | 240_fast___ |
| 241 | links - link - post - dude - sorry | 295 | 241_links_link_post_dude |
| 242 | old - process - sorry - late - complete | 293 | 242_old_process_sorry_late |
| 243 | cool - night - sir - friend - crazy | 292 | 243_cool_night_sir_friend |
| 244 | value - questions - low - price - liquidity | 292 | 244_value_questions_low_price |
| 245 | post - thing - online - month - page | 291 | 245_post_thing_online_month |
| 246 | late - step - plans - email - reward | 291 | 246_late_step_plans_email |
| 247 | work - yesterday - times - stuff - reason | 290 | 247_work_yesterday_times_stuff |
| 248 | home - minutes - hour - hours - months | 290 | 248_home_minutes_hour_hours |
| 249 | games - technology - game - blockchain - listing | 290 | 249_games_technology_game_blockchain |
| 250 | pepe - attention - buy - couple - stuff | 290 | 250_pepe_attention_buy_couple |
| 251 | power - trade - happy - company - withdrawal | 290 | 251_power_trade_happy_company |
| 252 | plans - fun - plan - happy - end | 288 | 252_plans_fun_plan_happy |
| 253 | online - legit - power - easy - extra | 287 | 253_online_legit_power_easy |
| 254 | buddy - special - yesterday - guess - wrong | 287 | 254_buddy_special_yesterday_guess |
| 255 | fine - thanks - mate - buddy - good | 287 | 255_fine_thanks_mate_buddy |
| 256 | stuff - power - potential - info - team | 286 | 256_stuff_power_potential_info |
| 257 | perfect - info - low - dm - team | 286 | 257_perfect_info_low_dm |
| 258 | projects - project - research - situation - lot | 286 | 258_projects_project_research_situation |
| 259 | version - update - latest - feature - control | 284 | 259_version_update_latest_feature |
| 260 | bit - buddy - sorry - little - work | 284 | 260_bit_buddy_sorry_little |
| 261 | success - game - community - best - | 284 | 261_success_game_community_best |
| 262 | market - - - - | 284 | 262_market___ |
| 263 | holders - term - strong - small - number | 282 | 263_holders_term_strong_small |
| 264 | public - means - fact - research - problems | 282 | 264_public_means_fact_research |
| 265 | run - awesome - simple - near - year | 282 | 265_run_awesome_simple_near |
| 266 | pls - half - price - current - better | 281 | 266_pls_half_price_current |
| 267 | reward - example - higher - site - rewards | 281 | 267_reward_example_higher_site |
| 268 | trust - difference - thought - words - deal | 281 | 268_trust_difference_thought_words |
| 269 | assets - longer - value - current - control | 280 | 269_assets_longer_value_current |
| 270 | day - dm - - - | 280 | 270_day_dm__ |
| 271 | country - friend - visit - brother - friends | 280 | 271_country_friend_visit_brother |
| 272 | place - visit - nice - cool - brother | 279 | 272_place_visit_nice_cool |
| 273 | search - process - cryptocurrency - investing - life | 279 | 273_search_process_cryptocurrency_investing |
| 274 | care - info - best - time - | 278 | 274_care_info_best_time |
| 275 | excited - power - search - times - trade | 277 | 275_excited_power_search_times |
| 276 | features - wallet - security - transactions - cryptocurrency | 277 | 276_features_wallet_security_transactions |
| 277 | crazy - accounts - ready - month - days | 276 | 277_crazy_accounts_ready_month |
| 278 | day - week - awesome - buddy - morning | 276 | 278_day_week_awesome_buddy |
| 279 | solution - rate - accounts - month - right | 275 | 279_solution_rate_accounts_month |
| 280 | listing - update - moment - data - chart | 275 | 280_listing_update_moment_data |
| 281 | address - correct - wrong - list - change | 274 | 281_address_correct_wrong_list |
| 282 | launch - product - release - announcement - close | 274 | 282_launch_product_release_announcement |
| 283 | community - group - discord - specific - guy | 273 | 283_community_group_discord_specific |
| 284 | chat - bit - mind - personal - nice | 271 | 284_chat_bit_mind_personal |
| 285 | fees - run - site - point - stuff | 271 | 285_fees_run_site_point |
| 286 | payment - - - - | 271 | 286_payment___ |
| 287 | longer - sense - value - country - moment | 270 | 287_longer_sense_value_country |
| 288 | ethereum - projects - tbh - solution - recent | 270 | 288_ethereum_projects_tbh_solution |
| 289 | latest - change - build - specific - difference | 270 | 289_latest_change_build_specific |
| 290 | pool - higher - current - rate - wonder | 269 | 290_pool_higher_current_rate |
| 291 | free - stuff - shit - job - version | 268 | 291_free_stuff_shit_job |
| 292 | bit - near - past - lot - moment | 268 | 292_bit_near_past_lot |
| 293 | holders - coin - value - coins - technology | 268 | 293_holders_coin_value_coins |
| 294 | fee - example - swap - contract - network | 268 | 294_fee_example_swap_contract |
| 295 | sell - article - longer - interesting - development | 267 | 295_sell_article_longer_interesting |
| 296 | service - month - info - marketing - offer | 267 | 296_service_month_info_marketing |
| 297 | single - web - plan - access - version | 267 | 297_single_web_plan_access |
| 298 | digital - exchange - token - - | 266 | 298_digital_exchange_token_ |
| 299 | thing - sorry - glad - dude - point | 266 | 299_thing_sorry_glad_dude |
| 300 | number - set - tbh - difference - user | 266 | 300_number_set_tbh_difference |
| 301 | note - update - global - phone - release | 264 | 301_note_update_global_phone |
| 302 | brother - morning - mate - buddy - welcome | 264 | 302_brother_morning_mate_buddy |
| 303 | day - hope - nice - great - mate | 263 | 303_day_hope_nice_great |
| 304 | awesome - successful - works - amazing - cryptocurrency | 262 | 304_awesome_successful_works_amazing |
| 305 | extra - lot - deposit - online - profits | 261 | 305_extra_lot_deposit_online |
| 306 | action - potential - success - community - best | 259 | 306_action_potential_success_community |
| 307 | face - deal - days - video - month | 258 | 307_face_deal_days_video |
| 308 | date - specific - announcement - special - updates | 257 | 308_date_specific_announcement_special |
| 309 | price - talk - chat - discord - group | 257 | 309_price_talk_chat_discord |
| 310 | happening - couple - event - massive - hand | 256 | 310_happening_couple_event_massive |
| 311 | chat - previous - happening - kinda - shit | 256 | 311_chat_previous_happening_kinda |
| 312 | ppl - person - head - face - little | 255 | 312_ppl_person_head_face |
| 313 | year - likely - longer - positive - close | 255 | 313_year_likely_longer_positive |
| 314 | certain - extra - lot - deposit - experience | 255 | 314_certain_extra_lot_deposit |
| 315 | technology - web - mate - job - platform | 255 | 315_technology_web_mate_job |
| 316 | phone - hand - max - old - half | 254 | 316_phone_hand_max_old |
| 317 | half - couple - guy - earlier - break | 254 | 317_half_couple_guy_earlier |
| 318 | chart - current - higher - early - lol | 254 | 318_chart_current_higher_early |
| 319 | bitcoin - day - transaction - happy - btc | 253 | 319_bitcoin_day_transaction_happy |
| 320 | account - email - process - face - help | 253 | 320_account_email_process_face |
| 321 | web - plan - version - month - info | 253 | 321_web_plan_version_month |
| 322 | fine - btw - ok - dear - nice | 253 | 322_fine_btw_ok_dear |
| 323 | short - mind - transactions - total - supply | 252 | 323_short_mind_transactions_total |
| 324 | business - action - group - success - ready | 250 | 324_business_action_group_success |
| 325 | volume - low - dude - higher - weeks | 249 | 325_volume_low_dude_higher |
| 326 | run - questions - address - update - network | 249 | 326_run_questions_address_update |
| 327 | active - message - available - account - dm | 249 | 327_active_message_available_account |
| 328 | message - previous - reply - longer - old | 248 | 328_message_previous_reply_longer |
| 329 | ca - option - sure - buddy - shit | 248 | 329_ca_option_sure_buddy |
| 330 | nice - glad - cool - dear - brother | 248 | 330_nice_glad_cool_dear |
| 331 | swap - buying - sell - pool - end | 247 | 331_swap_buying_sell_pool |
| 332 | link - protocol - version - chain - coin | 247 | 332_link_protocol_version_chain |
| 333 | contact - channel - available - - | 247 | 333_contact_channel_available_ |
| 334 | correct - accounts - info - offer - month | 246 | 334_correct_accounts_info_offer |
| 335 | chain - network - ecosystem - volume - solution | 246 | 335_chain_network_ecosystem_volume |
| 336 | info - members - offer - month - cool | 246 | 336_info_members_offer_month |
| 337 | brother - happy - - - | 245 | 337_brother_happy__ |
| 338 | easy - possible - hard - bro - likely | 244 | 338_easy_possible_hard_bro |
| 339 | spot - group - trading - okay - short | 244 | 339_spot_group_trading_okay |
| 340 | product - old - dude - nice - correct | 244 | 340_product_old_dude_nice |
| 341 | higher - quick - perfect - media - social | 244 | 341_higher_quick_perfect_media |
| 342 | market - year - thing - run - end | 243 | 342_market_year_thing_run |
| 343 | fake - real - ones - yesterday - person | 243 | 343_fake_real_ones_yesterday |
| 344 | rules - community - clear - group - change | 243 | 344_rules_community_clear_group |
| 345 | payment - fast - channel - good - dm | 242 | 345_payment_fast_channel_good |
| 346 | huge - high - available - - | 242 | 346_huge_high_available_ |
| 347 | difference - similar - different - better - question | 241 | 347_difference_similar_different_better |
| 348 | luck - tomorrow - fine - upcoming - nice | 241 | 348_luck_tomorrow_fine_upcoming |
| 349 | low - higher - high - previous - likely | 241 | 349_low_higher_high_previous |
| 350 | known - major - focus - cryptocurrency - massive | 241 | 350_known_major_focus_cryptocurrency |
| 351 | big - - - - | 240 | 351_big___ |
| 352 | guide - message - click - channel - access | 240 | 352_guide_message_click_channel |
| 353 | contact - person - email - potential - business | 240 | 353_contact_person_email_potential |
| 354 | nice - live - place - night - cool | 240 | 354_nice_live_place_night |
| 355 | list - smart - fact - damn - twitter | 239 | 355_list_smart_fact_damn |
| 356 | biggest - reason - fees - invest - money | 238 | 356_biggest_reason_fees_invest |
| 357 | success - key - mate - cause - future | 238 | 357_success_key_mate_cause |
| 358 | level - period - bit - active - minutes | 238 | 358_level_period_bit_active |
| 359 | answer - discord - questions - minutes - fun | 238 | 359_answer_discord_questions_minutes |
| 360 | funds - safe - situation - sorry - okay | 237 | 360_funds_safe_situation_sorry |
| 361 | country - nice - brother - situation - dear | 237 | 361_country_nice_brother_situation |
| 362 | pay - available - rate - app - deal | 237 | 362_pay_available_rate_app |
| 363 | alot - earning - place - best - platform | 237 | 363_alot_earning_place_best |
| 364 | bad - works - win - better - lot | 237 | 364_bad_works_win_better |
| 365 | version - fix - latest - phone - option | 236 | 365_version_fix_latest_phone |
| 366 | research - mate - cool - buddy - project | 236 | 366_research_mate_cool_buddy |
| 367 | website - link - admin - message - help | 235 | 367_website_link_admin_message |
| 368 | control - true - public - opinion - ones | 235 | 368_control_true_public_opinion |
| 369 | account - problem - correct - issues - issue | 234 | 369_account_problem_correct_issues |
| 370 | okay - alright - night - stop - bad | 234 | 370_okay_alright_night_stop |
| 371 | open - questions - hand - face - sounds | 234 | 371_open_questions_hand_face |
| 372 | amazing - info - support - best - free | 234 | 372_amazing_info_support_best |
| 373 | phone - version - latest - step - current | 234 | 373_phone_version_latest_step |
| 374 | market - run - dude - year - upcoming | 233 | 374_market_run_dude_year |
| 375 | chain - development - building - shit - sense | 233 | 375_chain_development_building_shit |
| 376 | glad - step - little - fine - brother | 233 | 376_glad_step_little_fine |
| 377 | trust - info - level - high - today | 232 | 377_trust_info_level_high |
| 378 | help - kind - type - course - hand | 232 | 378_help_kind_type_course |
| 379 | video - tho - words - admin - wrong | 232 | 379_video_tho_words_admin |
| 380 | cause - weeks - deposit - personal - key | 231 | 380_cause_weeks_deposit_personal |
| 381 | links - payment - available - - | 230 | 381_links_payment_available_ |
| 382 | listing - announcement - date - news - updates | 229 | 382_listing_announcement_date_news |
| 383 | pm - members - happy - payment - channel | 229 | 383_pm_members_happy_payment |
| 384 | list - access - best - - | 229 | 384_list_access_best_ |
| 385 | reason - chat - dude - rules - normal | 229 | 385_reason_chat_dude_rules |
| 386 | short - long - chance - yesterday - small | 229 | 386_short_long_chance_yesterday |
| 387 | marketing - hours - contract - group - start | 229 | 387_marketing_hours_contract_group |
| 388 | worth - looks - crazy - shit - lot | 229 | 388_worth_looks_crazy_shit |
| 389 | fun - home - interesting - dude - fine | 229 | 389_fun_home_interesting_dude |
| 390 | damn - ppl - bro - sense - face | 227 | 390_damn_ppl_bro_sense |
| 391 | media - amazing - info - team - | 227 | 391_media_amazing_info_team |
| 392 | available - - - - | 227 | 392_available___ |
| 393 | point - sense - opinion - comment - damn | 226 | 393_point_sense_opinion_comment |
| 394 | message - media - social - page - payment | 226 | 394_message_media_social_page |
| 395 | release - version - open - public - weeks | 226 | 395_release_version_open_public |
| 396 | - - - - | 226 | 396____ |
| 397 | buddy - token - ama - ai - cool | 226 | 397_buddy_token_ama_ai |
| 398 | successful - financial - right - payment - investors | 226 | 398_successful_financial_right_payment |
| 399 | pls - need - ppl - ways - dude | 226 | 399_pls_need_ppl_ways |
| 400 | deal - exchange - utc - telegram - website | 226 | 400_deal_exchange_utc_telegram |
| 401 | message - available - great - money - | 225 | 401_message_available_great_money |
| 402 | guy - man - bro - shit - thought | 225 | 402_guy_man_bro_shit |
| 403 | hour - week - month - day - contact | 225 | 403_hour_week_month_day |
| 404 | needs - public - perfect - minimum - year | 225 | 404_needs_public_perfect_minimum |
| 405 | today - - - - | 224 | 405_today___ |
| 406 | page - big - - - | 224 | 406_page_big__ |
| 407 | phone - update - main - sir - issue | 223 | 407_phone_update_main_sir |
| 408 | similar - awesome - interesting - old - love | 223 | 408_similar_awesome_interesting_old |
| 409 | ecosystem - action - fact - option - ama | 223 | 409_ecosystem_action_fact_option |
| 410 | alot - higher - true - financial - services | 222 | 410_alot_higher_true_financial |
| 411 | case - security - hold - volume - total | 222 | 411_case_security_hold_volume |
| 412 | services - hours - service - channel - free | 222 | 412_services_hours_service_channel |
| 413 | hours - real - available - free - | 222 | 413_hours_real_available_free |
| 414 | market - crypto - cryptocurrency - major - past | 221 | 414_market_crypto_cryptocurrency_major |
| 415 | page - available - - - | 221 | 415_page_available__ |
| 416 | hour - times - service - hours - days | 221 | 416_hour_times_service_hours |
| 417 | true - night - day - crazy - sense | 221 | 417_true_night_day_crazy |
| 418 | trust - wallet - problem - issues - option | 220 | 418_trust_wallet_problem_issues |
| 419 | fact - positive - mind - work - possible | 220 | 419_fact_positive_mind_work |
| 420 | day - mate - dear - friend - alright | 220 | 420_day_mate_dear_friend |
| 421 | high - dm - - - | 220 | 421_high_dm__ |
| 422 | plans - process - people - investors - platform | 220 | 422_plans_process_people_investors |
| 423 | ecosystem - attention - kind - focus - stuff | 220 | 423_ecosystem_attention_kind_focus |
| 424 | focus - potential - strong - future - huge | 220 | 424_focus_potential_strong_future |
| 425 | cool - wrong - dude - tomorrow - couple | 219 | 425_cool_wrong_dude_tomorrow |
| 426 | rest - work - account - minimum - hard | 219 | 426_rest_work_account_minimum |
| 427 | action - link - user - games - links | 218 | 427_action_link_user_games |
| 428 | risk - control - chance - imo - potential | 218 | 428_risk_control_chance_imo |
| 429 | step - public - earlier - break - close | 218 | 429_step_public_earlier_break |
| 430 | ecosystem - ama - excited - protocol - ethereum | 217 | 430_ecosystem_ama_excited_protocol |
| 431 | post - channel - - - | 217 | 431_post_channel__ |
| 432 | success - turn - step - financial - risk | 216 | 432_success_turn_step_financial |
| 433 | main - old - rate - single - half | 216 | 433_main_old_rate_single |
| 434 | reply - personal - dude - yesterday - special | 216 | 434_reply_personal_dude_yesterday |
| 435 | links - possible - things - information - dm | 216 | 435_links_possible_things_information |
| 436 | job - course - sure - moment - special | 215 | 436_job_course_sure_moment |
| 437 | value - important - info - members - week | 215 | 437_value_important_info_members |
| 438 | perfect - complete - hard - update - token | 215 | 438_perfect_complete_hard_update |
| 439 | channel - - - - | 215 | 439_channel___ |
| 440 | reason - important - limited - investment - company | 214 | 440_reason_important_limited_investment |
| 441 | bot - channel - - - | 214 | 441_bot_channel__ |
| 442 | amazing - info - high - support - | 214 | 442_amazing_info_high_support |
| 443 | major - difference - close - accounts - bro | 214 | 443_major_difference_close_accounts |
| 444 | drop - sense - lol - specific - shit | 213 | 444_drop_sense_lol_specific |
| 445 | buddy - happy - mate - nice - great | 213 | 445_buddy_happy_mate_nice |
| 446 | space - research - fake - fact - simple | 212 | 446_space_research_fake_fact |
| 447 | guy - friend - cool - country - mean | 212 | 447_guy_friend_cool_country |
| 448 | perfect - potential - years - marketing - experience | 211 | 448_perfect_potential_years_marketing |
| 449 | listing - early - special - pretty - related | 211 | 449_listing_early_special_pretty |
| 450 | low - price - crypto - daily - trading | 211 | 450_low_price_crypto_daily |
| 451 | matter - rules - clear - sounds - fair | 210 | 451_matter_rules_clear_sounds |
| 452 | spot - pool - listing - positive - current | 210 | 452_spot_pool_listing_positive |
| 453 | fine - phone - problems - previous - face | 210 | 453_fine_phone_problems_previous |
| 454 | features - feature - ecosystem - amazing - users | 210 | 454_features_feature_ecosystem_amazing |
| 455 | security - smart - report - contract - secure | 210 | 455_security_smart_report_contract |
| 456 | increase - massive - early - stuff - easy | 209 | 456_increase_massive_early_stuff |
| 457 | marketing - contact - form - email - head | 209 | 457_marketing_contact_form_email |
| 458 | past - rate - fair - accounts - days | 209 | 458_past_rate_fair_accounts |
| 459 | stuff - ones - form - certain - hand | 208 | 459_stuff_ones_form_certain |
| 460 | fee - fees - high - pls - cause | 208 | 460_fee_fees_high_pls |
| 461 | active - - - - | 207 | 461_active___ |
| 462 | usdt - withdraw - address - problem - wallet | 207 | 462_usdt_withdraw_address_problem |
| 463 | code - links - special - friends - ways | 207 | 463_code_links_special_friends |
| 464 | long - hour - minutes - weeks - months | 205 | 464_long_hour_minutes_weeks |
| 465 | ready - business - care - create - level | 205 | 465_ready_business_care_create |
| 466 | crazy - dude - cool - clear - couple | 205 | 466_crazy_dude_cool_clear |
| 467 | release - likely - feature - normal - example | 205 | 467_release_likely_feature_normal |
| 468 | info - years - fast - - | 204 | 468_info_years_fast_ |
| 469 | end - way - bitcoin - company - best | 204 | 469_end_way_bitcoin_company |
| 470 | spot - great - free - - | 204 | 470_spot_great_free_ |
| 471 | care - damn - dude - opinion - words | 203 | 471_care_damn_dude_opinion |
| 472 | page - channel - available - day - free | 203 | 472_page_channel_available_day |
| 473 | blockchain - development - smart - web - experience | 203 | 473_blockchain_development_smart_web |
| 474 | dev - coins - shit - sense - total | 202 | 474_dev_coins_shit_sense |
| 475 | sure - mate - comment - awesome - interesting | 201 | 475_sure_mate_comment_awesome |
| 476 | position - network - free - best - new | 201 | 476_position_network_free_best |
| 477 | reply - dm - buying - attention - help | 200 | 477_reply_dm_buying_attention |
| 478 | phone - guy - answer - important - shit | 199 | 478_phone_guy_answer_important |
| 479 | week - weeks - upcoming - great - ready | 199 | 479_week_weeks_upcoming_great |
| 480 | safe - - - - | 198 | 480_safe___ |
| 481 | higher - latest - likely - share - smart | 198 | 481_higher_latest_likely_share |
| 482 | excited - coin - chart - buddy - future | 197 | 482_excited_coin_chart_buddy |
| 483 | plans - process - easy - extra - profits | 197 | 483_plans_process_easy_extra |
| 484 | ecosystem - token - supply - clear - benefits | 197 | 484_ecosystem_token_supply_clear |
| 485 | nice - cool - home - alright - kinda | 197 | 485_nice_cool_home_alright |
| 486 | sir - exchange - trade - withdrawal - deposit | 197 | 486_sir_exchange_trade_withdrawal |
| 487 | sign - services - profits - payment - investors | 196 | 487_sign_services_profits_payment |
| 488 | close - ta - chart - crazy - previous | 196 | 488_close_ta_chart_crazy |
| 489 | run - half - guy - sounds - fact | 196 | 489_run_half_guy_sounds |
| 490 | phone - pay - trust - official - payment | 195 | 490_phone_pay_trust_official |
| 491 | safe - sorry - brother - better - buddy | 195 | 491_safe_sorry_brother_better |
| 492 | network - number - value - product - business | 195 | 492_network_number_value_product |
| 493 | thought - mate - lol - thing - buddy | 195 | 493_thought_mate_lol_thing |
| 494 | perfect - potential - info - experience - | 194 | 494_perfect_potential_info_experience |
| 495 | security - secure - opinion - issue - important | 193 | 495_security_secure_opinion_issue |
| 496 | dude - mate - man - yesterday - past | 193 | 496_dude_mate_man_yesterday |
| 497 | high - page - value - network - | 192 | 497_high_page_value_network |
| 498 | position - higher - pay - free - best | 192 | 498_position_higher_pay_free |
| 499 | listing - token - exchange - attention - project | 192 | 499_listing_token_exchange_attention |
| 500 | special - related - discord - development - blockchain | 192 | 500_special_related_discord_development |
| 501 | chat - certain - global - cause - alot | 192 | 501_chat_certain_global_cause |
| 502 | similar - face - couple - ones - brother | 192 | 502_similar_face_couple_ones |
| 503 | option - step - life - successful - success | 192 | 503_option_step_life_successful |
| 504 | ecosystem - web - private - secure - personal | 191 | 504_ecosystem_web_private_secure |
| 505 | weeks - date - sir - announcement - month | 190 | 505_weeks_date_sir_announcement |
| 506 | guy - man - btw - attention - half | 190 | 506_guy_man_btw_attention |
| 507 | course - access - telegram - free - start | 190 | 507_course_access_telegram_free |
| 508 | possible - cool - times - way - | 190 | 508_possible_cool_times_way |
| 509 | late - early - chance - minutes - point | 189 | 509_late_early_chance_minutes |
| 510 | tbh - ways - cool - plan - profits | 189 | 510_tbh_ways_cool_plan |
| 511 | normal - problems - phone - solution - old | 189 | 511_normal_problems_phone_solution |
| 512 | words - hand - interesting - specific - love | 189 | 512_words_hand_interesting_specific |
| 513 | video - pay - chat - hours - fast | 188 | 513_video_pay_chat_hours |
| 514 | sounds - bit - questions - normal - better | 188 | 514_sounds_bit_questions_normal |
| 515 | reason - important - success - funds - safe | 188 | 515_reason_important_success_funds |
| 516 | list - event - announcement - sir - week | 188 | 516_list_event_announcement_sir |
| 517 | release - holders - game - important - supply | 187 | 517_release_holders_game_important |
| 518 | simple - time - - - | 187 | 518_simple_time__ |
| 519 | dm - message - dear - contact - sorry | 187 | 519_dm_message_dear_contact |
| 520 | game - special - feature - second - little | 186 | 520_game_special_feature_second |
| 521 | option - normal - main - correct - problem | 186 | 521_option_normal_main_correct |
| 522 | sense - country - cool - half - break | 186 | 522_sense_country_cool_half |
| 523 | fine - great - thanks - buddy - mate | 185 | 523_fine_great_thanks_buddy |
| 524 | links - video - year - available - payment | 185 | 524_links_video_year_available |
| 525 | similar - business - special - correct - old | 185 | 525_similar_business_special_correct |
| 526 | days - message - fast - available - | 185 | 526_days_message_fast_available |
| 527 | quick - plans - transaction - deposit - easy | 185 | 527_quick_plans_transaction_deposit |
| 528 | protocol - assets - rate - liquidity - small | 185 | 528_protocol_assets_rate_liquidity |
| 529 | active - earning - info - withdrawal - money | 184 | 529_active_earning_info_withdrawal |
| 530 | glad - believe - key - little - financial | 184 | 530_glad_believe_key_little |
| 531 | kinda - hand - person - couple - position | 183 | 531_kinda_hand_person_couple |
| 532 | mate - research - buddy - project - interesting | 183 | 532_mate_research_buddy_project |
| 533 | - - - - | 182 | 533____ |
| 534 | cool - previous - guy - face - tbh | 181 | 534_cool_previous_guy_face |
| 535 | deal - account - positive - accounts - daily | 180 | 535_deal_account_positive_accounts |
| 536 | admin - dude - rules - sorry - comment | 180 | 536_admin_dude_rules_sorry |
| 537 | fine - glad - brother - bro - dear | 180 | 537_fine_glad_brother_bro |
| 538 | hour - bot - telegram - day - | 179 | 538_hour_bot_telegram_day |
| 539 | development - interesting - recent - latest - exciting | 179 | 539_development_interesting_recent_latest |
| 540 | shit - coins - stuff - biggest - luck | 179 | 540_shit_coins_stuff_biggest |
| 541 | guys - free - money - work - | 179 | 541_guys_free_money_work |
| 542 | feature - major - dev - game - face | 178 | 542_feature_major_dev_game |
| 543 | type - big - account - available - | 177 | 543_type_big_account_available |
| 544 | mate - man - event - exchange - rewards | 177 | 544_mate_man_event_exchange |
| 545 | spot - trading - position - difference - trade | 177 | 545_spot_trading_position_difference |
| 546 | happy - friend - special - year - family | 177 | 546_happy_friend_special_year |
| 547 | close - likely - sign - upcoming - term | 177 | 547_close_likely_sign_upcoming |
| 548 | holders - benefits - extra - buying - appreciate | 177 | 548_holders_benefits_extra_buying |
| 549 | benefits - fee - deposit - investment - withdrawal | 177 | 549_benefits_fee_deposit_investment |
| 550 | week - weeks - night - close - nice | 175 | 550_week_weeks_night_close |
| 551 | media - social - message - available - free | 175 | 551_media_social_message_available |
| 552 | special - buddy - day - mate - order | 175 | 552_special_buddy_day_mate |
| 553 | action - process - success - investors - right | 175 | 553_action_process_success_investors |
| 554 | today - thanks - - - | 175 | 554_today_thanks__ |
| 555 | date - video - contact - happy - channel | 175 | 555_date_video_contact_happy |
| 556 | similar - hand - investors - ecosystem - normal | 174 | 556_similar_hand_investors_ecosystem |
| 557 | turn - period - short - risk - best | 174 | 557_turn_period_short_risk |
| 558 | protocol - transactions - wonder - number - liquidity | 174 | 558_protocol_transactions_wonder_number |
| 559 | public - things - group - company - best | 174 | 559_public_things_group_company |
| 560 | - - - - | 173 | 560____ |
| 561 | positive - financial - development - better - extra | 173 | 561_positive_financial_development_better |
| 562 | pepe - coin - recent - coins - fun | 173 | 562_pepe_coin_recent_coins |
| 563 | live - app - sense - address - half | 172 | 563_live_app_sense_address |
| 564 | difference - excited - fee - access - withdrawal | 172 | 564_difference_excited_fee_access |
| 565 | swap - utc - pm - details - time | 172 | 565_swap_utc_pm_details |
| 566 | different - plans - best - platform - right | 172 | 566_different_plans_best_platform |
| 567 | games - questions - buddy - exciting - user | 172 | 567_games_questions_buddy_exciting |
| 568 | happy - past - weeks - man - cool | 172 | 568_happy_past_weeks_man |
| 569 | tomorrow - action - family - issue - successful | 171 | 569_tomorrow_action_family_issue |
| 570 | key - legit - stuff - damn - works | 171 | 570_key_legit_stuff_damn |
| 571 | fair - launch - true - financial - website | 171 | 571_fair_launch_true_financial |
| 572 | needs - plan - offer - market - free | 171 | 572_needs_plan_offer_market |
| 573 | excited - exciting - brother - pretty - crazy | 171 | 573_excited_exciting_brother_pretty |
| 574 | simple - supply - official - website - telegram | 170 | 574_simple_supply_official_website |
| 575 | country - situation - buddy - guy - opinion | 170 | 575_country_situation_buddy_guy |
| 576 | rate - business - deal - legit - interested | 170 | 576_rate_business_deal_legit |
| 577 | guys - friend - man - dear - fine | 170 | 577_guys_friend_man_dear |
| 578 | experience - plus - build - running - problems | 170 | 578_experience_plus_build_running |
| 579 | report - past - guy - perfect - thanks | 170 | 579_report_past_guy_perfect |
| 580 | cool - nice - tho - ones - major | 169 | 580_cool_nice_tho_ones |
| 581 | supply - tokens - token - plan - total | 169 | 581_supply_tokens_token_plan |
| 582 | updates - details - pool - share - trading | 169 | 582_updates_details_pool_share |
| 583 | message - available - today - dm - period | 169 | 583_message_available_today_dm |
| 584 | connect - fine - problems - issue - issues | 169 | 584_connect_fine_problems_issue |
| 585 | form - yesterday - clear - okay - question | 168 | 585_form_yesterday_clear_okay |
| 586 | year - dude - course - shit - smart | 168 | 586_year_dude_course_shit |
| 587 | positive - world - love - ways - luck | 168 | 587_positive_world_love_ways |
| 588 | nice - face - global - country - interested | 168 | 588_nice_face_global_country |
| 589 | game - site - reply - issues - problem | 168 | 589_game_site_reply_issues |
| 590 | list - spot - later - glad - event | 167 | 590_list_spot_later_glad |
| 591 | previous - year - new - term - guess | 167 | 591_previous_year_new_term |
| 592 | write - words - stuff - sense - hand | 167 | 592_write_words_stuff_sense |
| 593 | tho - mate - true - need - friend | 167 | 593_tho_mate_true_need |
| 594 | key - private - public - correct - hand | 167 | 594_key_private_public_correct |
| 595 | legit - end - extra - online - way | 167 | 595_legit_end_extra_online |
| 596 | successful - change - financial - payment - investors | 167 | 596_successful_change_financial_payment |
| 597 | dev - idea - correct - number - difference | 166 | 597_dev_idea_correct_number |
| 598 | ethereum - end - chain - site - problems | 165 | 598_ethereum_end_chain_site |
| 599 | bot - guy - word - lol - correct | 165 | 599_bot_guy_word_lol |
| 600 | exchange - upcoming - ecosystem - technology - updates | 165 | 600_exchange_upcoming_ecosystem_technology |
| 601 | head - couple - okay - buddy - weeks | 165 | 601_head_couple_okay_buddy |
| 602 | data - security - technology - features - ones | 164 | 602_data_security_technology_features |
| 603 | features - feature - unique - cool - kind | 164 | 603_features_feature_unique_cool |
| 604 | wrong - shit - guy - mate - brother | 164 | 604_wrong_shit_guy_mate |
| 605 | event - special - guys - interesting - amazing | 163 | 605_event_special_guys_interesting |
| 606 | brother - head - late - care - little | 163 | 606_brother_head_late_care |
| 607 | lol - haha - answer - situation - question | 163 | 607_lol_haha_answer_situation |
| 608 | tomorrow - little - better - future - opportunity | 162 | 608_tomorrow_little_better_future |
| 609 | - - - - | 162 | 609____ |
| 610 | yesterday - dude - interesting - fact - tho | 162 | 610_yesterday_dude_interesting_fact |
| 611 | key - support - positive - play - price | 161 | 611_key_support_positive_play |
| 612 | available - page - good - questions - perfect | 160 | 612_available_page_good_questions |
| 613 | spot - real - page - guys - channel | 160 | 613_spot_real_page_guys |
| 614 | simple - time - new - - | 160 | 614_simple_time_new_ |
| 615 | point - fun - ways - option - increase | 160 | 615_point_fun_ways_option |
| 616 | protocol - solution - comment - features - ecosystem | 160 | 616_protocol_solution_comment_features |
| 617 | step - successful - thing - life - money | 160 | 617_step_successful_thing_life |
| 618 | active - big - today - - | 159 | 618_active_big_today_ |
| 619 | words - payment - blockchain - wallet - profit | 159 | 619_words_payment_blockchain_wallet |
| 620 | focus - main - online - thing - easy | 159 | 620_focus_main_online_thing |
| 621 | reason - important - limited - investment - company | 159 | 621_reason_important_limited_investment |
| 622 | personal - mate - opinion - bro - buddy | 159 | 622_personal_mate_opinion_bro |
| 623 | pool - utc - fair - second - yesterday | 159 | 623_pool_utc_fair_second |
| 624 | months - dude - yesterday - weeks - shit | 159 | 624_months_dude_yesterday_weeks |
| 625 | info - marketing - - - | 159 | 625_info_marketing__ |
| 626 | lose - extra - experience - best - platform | 158 | 626_lose_extra_experience_best |
| 627 | happy - day - special - family - morning | 158 | 627_happy_day_special_family |
| 628 | fair - situation - games - months - specific | 158 | 628_fair_situation_games_months |
| 629 | bitcoin - past - major - drop - similar | 158 | 629_bitcoin_past_major_drop |
| 630 | job - work - kinda - mean - kind | 158 | 630_job_work_kinda_mean |
| 631 | rate - year - wonder - clear - country | 158 | 631_rate_year_wonder_clear |
| 632 | difference - issues - strong - digital - true | 157 | 632_difference_issues_strong_digital |
| 633 | earlier - yesterday - sense - morning - night | 157 | 633_earlier_yesterday_sense_morning |
| 634 | old - nice - fine - sorry - cool | 157 | 634_old_nice_fine_sorry |
| 635 | services - message - channel - telegram - projects | 157 | 635_services_message_channel_telegram |
| 636 | risk - capital - life - investment - | 156 | 636_risk_capital_life_investment |
| 637 | level - example - account - tomorrow - future | 156 | 637_level_example_account_tomorrow |
| 638 | active - nice - similar - mate - year | 156 | 638_active_nice_similar_mate |
| 639 | brother - stuff - things - ones - cool | 156 | 639_brother_stuff_things_ones |
| 640 | guide - course - selling - months - complete | 156 | 640_guide_course_selling_months |
| 641 | interested - address - email - profits - blockchain | 155 | 641_interested_address_email_profits |
| 642 | update - version - specific - announcement - official | 155 | 642_update_version_specific_announcement |
| 643 | step - plans - cryptocurrency - fast - life | 154 | 643_step_plans_cryptocurrency_fast |
| 644 | post - bro - situation - question - stuff | 154 | 644_post_bro_situation_question |
| 645 | leave - reason - okay - mate - correct | 154 | 645_leave_reason_okay_mate |
| 646 | near - potential - little - development - rate | 154 | 646_near_potential_little_development |
| 647 | period - months - release - article - claim | 154 | 647_period_months_release_article |
| 648 | future - near - clear - thinking - upcoming | 154 | 648_future_near_clear_thinking |
| 649 | details - information - bit - mate - question | 154 | 649_details_information_bit_mate |
| 650 | break - space - love - kind - matter | 154 | 650_break_space_love_kind |
| 651 | head - long - short - look - face | 153 | 651_head_long_short_look |
| 652 | super - technology - bit - nice - longer | 152 | 652_super_technology_bit_nice |
| 653 | funds - assets - exchange - announcement - product | 152 | 653_funds_assets_exchange_announcement |
| 654 | position - short - open - low - spot | 151 | 654_position_short_open_low |
| 655 | guy - sense - meet - interesting - upcoming | 151 | 655_guy_sense_meet_interesting |
| 656 | ecosystem - exciting - welcome - updates - development | 151 | 656_ecosystem_exciting_welcome_updates |
| 657 | major - play - special - upcoming - chance | 151 | 657_major_play_special_upcoming |
| 658 | difference - search - tbh - stuff - building | 151 | 658_difference_search_tbh_stuff |
| 659 | real - good - platform - - | 150 | 659_real_good_platform_ |
| 660 | single - shit - job - bro - care | 150 | 660_single_shit_job_bro |
| 661 | phone - issue - brother - normal - type | 150 | 661_phone_issue_brother_normal |
| 662 | glad - nice - mate - happy - fine | 150 | 662_glad_nice_mate_happy |
| 663 | research - fee - projects - free - tokens | 150 | 663_research_fee_projects_free |
| 664 | video - post - chat - page - hours | 150 | 664_video_post_chat_page |
| 665 | win - hope - brother - sir - mate | 149 | 665_win_hope_brother_sir |
| 666 | difference - price - correct - sense - opinion | 149 | 666_difference_price_correct_sense |
| 667 | bot - sense - cool - fun - channel | 149 | 667_bot_sense_cool_fun |
| 668 | message - number - situation - reply - crazy | 149 | 668_message_number_situation_reply |
| 669 | investing - successful - real - company - great | 149 | 669_investing_successful_real_company |
| 670 | nft - feature - complete - glad - normal | 149 | 670_nft_feature_complete_glad |
| 671 | smart - big - major - position - buddy | 148 | 671_smart_big_major_position |
| 672 | reach - huge - media - social - accounts | 148 | 672_reach_huge_media_social |
| 673 | stuff - shit - ones - single - night | 148 | 673_stuff_shit_ones_single |
| 674 | page - active - fast - available - | 147 | 674_page_active_fast_available |
| 675 | guy - point - fair - head - country | 147 | 675_guy_point_fair_head |
| 676 | game - shit - head - product - update | 147 | 676_game_shit_head_product |
| 677 | world - investment - company - people - best | 147 | 677_world_investment_company_people |
| 678 | country - crypto - interesting - space - rules | 147 | 678_country_crypto_interesting_space |
| 679 | rate - minimum - check - account - free | 146 | 679_rate_minimum_check_account |
| 680 | stuff - brother - lots - kinda - lol | 146 | 680_stuff_brother_lots_kinda |
| 681 | interested - major - solution - interesting - listing | 146 | 681_interested_major_solution_interesting |
| 682 | huge - video - chat - page - payment | 146 | 682_huge_video_chat_page |
| 683 | key - bit - word - brother - little | 145 | 683_key_bit_word_brother |
| 684 | wrong - correct - future - fact - positive | 145 | 684_wrong_correct_future_fact |
| 685 | month - end - date - year - tbh | 145 | 685_month_end_date_year |
| 686 | miss - okay - fine - glad - friend | 145 | 686_miss_okay_fine_glad |
| 687 | post - payment - fast - today - free | 145 | 687_post_payment_fast_today |
| 688 | list - wonder - man - bro - long | 145 | 688_list_wonder_man_bro |
| 689 | users - high - big - time - | 144 | 689_users_high_big_time |
| 690 | half - term - lose - crazy - fine | 144 | 690_half_term_lose_crazy |
| 691 | payment - app - crypto - solution - web | 144 | 691_payment_app_crypto_solution |
| 692 | sell - high - normal - rate - dm | 144 | 692_sell_high_normal_rate |
| 693 | open - earlier - announcement - yesterday - weeks | 144 | 693_open_earlier_announcement_yesterday |
| 694 | data - chat - user - information - users | 144 | 694_data_chat_user_information |
| 695 | fine - mate - great - man - glad | 143 | 695_fine_mate_great_man |
| 696 | building - things - product - stuff - single | 143 | 696_building_things_product_stuff |
| 697 | term - fun - space - long - project | 143 | 697_term_fun_space_long |
| 698 | fast - - - - | 143 | 698_fast___ |
| 699 | protocol - coin - difference - token - blockchain | 143 | 699_protocol_coin_difference_token |
| 700 | recent - example - ethereum - correct - wrong | 143 | 700_recent_example_ethereum_correct |
| 701 | token - claim - tokens - imo - deposit | 143 | 701_token_claim_tokens_imo |
| 702 | contact - post - page - payment - | 143 | 702_contact_post_page_payment |
| 703 | friends - reward - thanks - investors - new | 143 | 703_friends_reward_thanks_investors |
| 704 | question - dude - opinion - comment - hard | 142 | 704_question_dude_opinion_comment |
| 705 | event - weeks - correct - head - cool | 142 | 705_event_weeks_correct_head |
| 706 | announcement - luck - positive - clear - job | 142 | 706_announcement_luck_positive_clear |
| 707 | things - moment - tbh - true - high | 142 | 707_things_moment_tbh_true |
| 708 | alot - couple - man - course - sure | 141 | 708_alot_couple_man_course |
| 709 | face - little - investing - thanks - investment | 141 | 709_face_little_investing_thanks |
| 710 | multiple - perfect - times - sure - website | 141 | 710_multiple_perfect_times_sure |
| 711 | higher - dude - price - market - fact | 141 | 711_higher_dude_price_market |
| 712 | close - hand - super - higher - bought | 140 | 712_close_hand_super_higher |
| 713 | half - shit - opinion - product - previous | 140 | 713_half_shit_opinion_product |
| 714 | stuff - similar - thing - guy - cool | 140 | 714_stuff_similar_thing_guy |
| 715 | correct - data - coin - sense - volume | 140 | 715_correct_data_coin_sense |
| 716 | date - period - multiple - yesterday - latest | 140 | 716_date_period_multiple_yesterday |
| 717 | google - code - phone - app - key | 139 | 717_google_code_phone_app |
| 718 | security - home - kind - bot - development | 139 | 718_security_home_kind_bot |
| 719 | old - years - half - correct - man | 139 | 719_old_years_half_correct |
| 720 | person - shit - words - bit - sense | 139 | 720_person_shit_words_bit |
| 721 | end - course - long - mate - comment | 138 | 721_end_course_long_mate |
| 722 | correct - example - number - dude - couple | 138 | 722_correct_example_number_dude |
| 723 | head - specific - questions - answer - super | 137 | 723_head_specific_questions_answer |
| 724 | bit - deal - weeks - holders - admin | 137 | 724_bit_deal_weeks_holders |
| 725 | feature - problems - near - crypto - solution | 137 | 725_feature_problems_near_crypto |
| 726 | legit - home - benefits - success - experience | 137 | 726_legit_home_benefits_success |
| 727 | months - page - amazing - fast - week | 136 | 727_months_page_amazing_fast |
| 728 | moment - job - exchange - morning - community | 136 | 728_moment_job_exchange_morning |
| 729 | possible - group - time - - | 136 | 729_possible_group_time_ |
| 730 | comment - family - twitter - let - support | 136 | 730_comment_family_twitter_let |
| 731 | huge - amazing - great - time - best | 136 | 731_huge_amazing_great_time |
| 732 | bro - wan - yesterday - person - mate | 136 | 732_bro_wan_yesterday_person |
| 733 | nft - details - option - technology - product | 136 | 733_nft_details_option_technology |
| 734 | hear - glad - fine - nice - alright | 136 | 734_hear_glad_fine_nice |
| 735 | home - guy - fun - bro - nice | 135 | 735_home_guy_fun_bro |
| 736 | happy - times - bit - alright - sorry | 135 | 736_happy_times_bit_alright |
| 737 | dude - bro - true - man - point | 135 | 737_dude_bro_true_man |
| 738 | hello - reply - question - times - little | 135 | 738_hello_reply_question_times |
| 739 | links - ask - twitter - multiple - best | 135 | 739_links_ask_twitter_multiple |
| 740 | smart - people - man - hard - dude | 135 | 740_smart_people_man_hard |
| 741 | article - read - interesting - details - kind | 134 | 741_article_read_interesting_details |
| 742 | phone - ecosystem - problems - cause - similar | 134 | 742_phone_ecosystem_problems_cause |
| 743 | matter - higher - swap - dm - minimum | 134 | 743_matter_higher_swap_dm |
| 744 | available - big - dm - - | 134 | 744_available_big_dm_ |
| 745 | control - difference - ways - crazy - moment | 134 | 745_control_difference_ways_crazy |
| 746 | news - ai - ethereum - volume - crypto | 134 | 746_news_ai_ethereum_volume |
| 747 | games - technology - example - playing - feature | 134 | 747_games_technology_example_playing |
| 748 | long - longer - weeks - years - months | 134 | 748_long_longer_weeks_years |
| 749 | participate - ones - interesting - luck - sure | 133 | 749_participate_ones_interesting_luck |
| 750 | win - risk - level - official - daily | 133 | 750_win_risk_level_official |
| 751 | ready - channel - - - | 133 | 751_ready_channel__ |
| 752 | active - binance - payment - big - channel | 133 | 752_active_binance_payment_big |
| 753 | volume - trading - trade - total - sell | 133 | 753_volume_trading_trade_total |
| 754 | profits - reach - strong - report - market | 133 | 754_profits_reach_strong_report |
| 755 | normal - hand - address - chain - btc | 133 | 755_normal_hand_address_chain |
| 756 | word - type - words - similar - country | 132 | 756_word_type_words_similar |
| 757 | awesome - capital - experience - join - platform | 132 | 757_awesome_capital_experience_join |
| 758 | article - word - sir - reason - update | 132 | 758_article_word_sir_reason |
| 759 | swap - token - tokens - normal - similar | 132 | 759_swap_token_tokens_normal |
| 760 | media - social - ready - high - page | 131 | 760_media_social_ready_high |
| 761 | technology - data - care - difference - example | 131 | 761_technology_data_care_difference |
| 762 | close - sense - year - break - tho | 131 | 762_close_sense_year_break |
| 763 | private - key - smart - report - services | 131 | 763_private_key_smart_report |
| 764 | rate - website - invest - person - best | 130 | 764_rate_website_invest_person |
| 765 | year - guy - head - situation - drop | 130 | 765_year_guy_head_situation |
| 766 | page - available - free - day - message | 130 | 766_page_available_free_day |
| 767 | playing - phone - games - bit - build | 130 | 767_playing_phone_games_bit |
| 768 | video - available - high - active - free | 130 | 768_video_available_high_active |
| 769 | words - word - bit - little - wonder | 130 | 769_words_word_bit_little |
| 770 | idea - bad - good - thought - nice | 129 | 770_idea_bad_good_thought |
| 771 | wrong - shit - bad - words - fact | 129 | 771_wrong_shit_bad_words |
| 772 | mind - thought - crazy - shit - head | 129 | 772_mind_thought_crazy_shit |
| 773 | phone - hand - running - update - special | 129 | 773_phone_hand_running_update |
| 774 | close - couple - number - dude - value | 129 | 774_close_couple_number_dude |
| 775 | excited - hand - fun - couple - yesterday | 128 | 775_excited_hand_fun_couple |
| 776 | media - social - limited - - | 128 | 776_media_social_limited_ |
| 777 | dude - open - similar - man - project | 128 | 777_dude_open_similar_man |
| 778 | sign - list - past - tho - trade | 128 | 778_sign_list_past_tho |
| 779 | chain - tokens - address - projects - solution | 128 | 779_chain_tokens_address_projects |
| 780 | pool - example - article - specific - higher | 127 | 780_pool_example_article_specific |
| 781 | mate - home - glad - luck - sorry | 127 | 781_mate_home_glad_luck |
| 782 | plans - buddy - question - kind - website | 127 | 782_plans_buddy_question_kind |
| 783 | video - version - chance - trust - success | 127 | 783_video_version_chance_trust |
| 784 | huge - fees - withdrawal - live - chat | 126 | 784_huge_fees_withdrawal_live |
| 785 | ways - excited - important - form - details | 126 | 785_ways_excited_important_form |
| 786 | quick - possible - amazing - info - dm | 126 | 786_quick_possible_amazing_info |
| 787 | lol - reply - fun - cool - damn | 126 | 787_lol_reply_fun_cool |
| 788 | correct - half - total - clear - past | 126 | 788_correct_half_total_clear |
| 789 | media - social - limited - services - info | 126 | 789_media_social_limited_services |
| 790 | talk - guy - sir - things - okay | 126 | 790_talk_guy_sir_things |
| 791 | interesting - sounds - stuff - attention - ones | 126 | 791_interesting_sounds_stuff_attention |
| 792 | twitter - weeks - control - main - dude | 125 | 792_twitter_weeks_control_main |
| 793 | email - reply - yesterday - weeks - address | 125 | 793_email_reply_yesterday_weeks |
| 794 | trading - interested - trade - learn - family | 125 | 794_trading_interested_trade_learn |
| 795 | biggest - giving - period - hour - profits | 125 | 795_biggest_giving_period_hour |
| 796 | interested - financial - profits - click - lot | 125 | 796_interested_financial_profits_click |
| 797 | plans - process - extra - trading - deposit | 125 | 797_plans_process_extra_trading |
| 798 | ai - data - power - awesome - services | 125 | 798_ai_data_power_awesome |
| 799 | place - extra - active - lot - company | 125 | 799_place_extra_active_lot |
| 800 | bot - ethereum - video - eth - exciting | 125 | 800_bot_ethereum_video_eth |
| 801 | interesting - cool - short - sorry - guy | 124 | 801_interesting_cool_short_sorry |
| 802 | correct - change - wrong - fine - user | 124 | 802_correct_change_wrong_fine |
| 803 | successful - power - exchange - way - money | 124 | 803_successful_power_exchange_way |
| 804 | version - phone - build - dev - old | 124 | 804_version_phone_build_dev |
| 805 | digital - blockchain - trust - protocol - assets | 123 | 805_digital_blockchain_trust_protocol |
| 806 | lose - crypto - risk - investing - research | 123 | 806_lose_crypto_risk_investing |
| 807 | mate - awesome - friend - crazy - correct | 123 | 807_mate_awesome_friend_crazy |
| 808 | protocol - multiple - similar - fair - example | 123 | 808_protocol_multiple_similar_fair |
| 809 | profit - online - support - investment - good | 123 | 809_profit_online_support_investment |
| 810 | digital - assets - control - financial - opinion | 123 | 810_digital_assets_control_financial |
| 811 | games - country - shit - game - world | 123 | 811_games_country_shit_game |
| 812 | usdt - interested - daily - example - number | 123 | 812_usdt_interested_daily_example |
| 813 | - - - - | 122 | 813____ |
| 814 | games - man - huge - mate - buddy | 122 | 814_games_man_huge_mate |
| 815 | message - big - available - free - | 122 | 815_message_big_available_free |
| 816 | follow - starting - comment - sir - man | 122 | 816_follow_starting_comment_sir |
| 817 | sorry - way - buddy - difference - mate | 122 | 817_sorry_way_buddy_difference |
| 818 | global - year - risk - high - days | 122 | 818_global_year_risk_high |
| 819 | user - live - type - exciting - good | 122 | 819_user_live_type_exciting |
| 820 | buy - shit - moment - issues - ethereum | 121 | 820_buy_shit_moment_issues |
| 821 | small - man - kinda - shit - dude | 121 | 821_small_man_kinda_shit |
| 822 | cool - friend - previous - tho - weeks | 121 | 822_cool_friend_previous_tho |
| 823 | technology - word - words - lol - brother | 121 | 823_technology_word_words_lol |
| 824 | spot - channel - available - - | 121 | 824_spot_channel_available_ |
| 825 | link - - - - | 121 | 825_link___ |
| 826 | members - secure - accounts - follow - money | 121 | 826_members_secure_accounts_follow |
| 827 | hand - excited - things - questions - note | 121 | 827_hand_excited_things_questions |
| 828 | bitcoin - holders - wallet - btc - | 120 | 828_bitcoin_holders_wallet_btc |
| 829 | interesting - known - business - company - cool | 120 | 829_interesting_known_business_company |
| 830 | digital - family - payment - access - bitcoin | 120 | 830_digital_family_payment_access |
| 831 | option - period - date - fee - problem | 120 | 831_option_period_date_fee |
| 832 | meet - guy - fun - interested - fair | 120 | 832_meet_guy_fun_interested |
| 833 | group - related - fine - global - support | 119 | 833_group_related_fine_global |
| 834 | yesterday - transaction - spot - couple - problem | 119 | 834_yesterday_transaction_spot_couple |
| 835 | period - minutes - dear - utc - announcement | 119 | 835_period_minutes_dear_utc |
| 836 | guide - course - selling - months - complete | 119 | 836_guide_course_selling_months |
| 837 | news - contract - past - dev - ecosystem | 118 | 837_news_contract_past_dev |
| 838 | code - correct - problem - block - user | 118 | 838_code_correct_problem_block |
| 839 | reason - buddy - specific - main - sure | 118 | 839_reason_buddy_specific_main |
| 840 | weeks - post - page - hours - channel | 118 | 840_weeks_post_page_hours |
| 841 | moment - friend - strong - dear - early | 118 | 841_moment_friend_strong_dear |
| 842 | list - couple - meet - live - real | 118 | 842_list_couple_meet_live |
| 843 | address - investing - wallet - best - platform | 118 | 843_address_investing_wallet_best |
| 844 | strong - dev - huge - community - marketing | 118 | 844_strong_dev_huge_community |
| 845 | website - normal - man - post - bit | 118 | 845_website_normal_man_post |
| 846 | utc - spot - questions - listing - main | 117 | 846_utc_spot_questions_listing |
| 847 | country - hour - different - ok - morning | 117 | 847_country_hour_different_ok |
| 848 | dude - man - list - spot - hold | 117 | 848_dude_man_list_spot |
| 849 | increase - bot - members - unique - telegram | 117 | 849_increase_bot_members_unique |
| 850 | build - date - follow - latest - official | 117 | 850_build_date_follow_latest |
| 851 | words - key - stuff - love - official | 117 | 851_words_key_stuff_love |
| 852 | live - chat - support - type - issue | 116 | 852_live_chat_support_type |
| 853 | btc - supply - level - close - yesterday | 116 | 853_btc_supply_level_close |
| 854 | answer - questions - example - words - word | 116 | 854_answer_questions_example_words |
| 855 | bad - watch - course - brother - good | 116 | 855_bad_watch_course_brother |
| 856 | personal - major - example - trade - funds | 116 | 856_personal_major_example_trade |
| 857 | attention - buddy - upcoming - thanks - tomorrow | 115 | 857_attention_buddy_upcoming_thanks |
| 858 | data - phone - app - previous - tho | 115 | 858_data_phone_app_previous |
| 859 | late - short - close - open - kinda | 115 | 859_late_short_close_open |
| 860 | issues - change - issue - clear - try | 115 | 860_issues_change_issue_clear |
| 861 | tomorrow - person - real - successful - services | 115 | 861_tomorrow_person_real_successful |
| 862 | specific - main - close - interesting - hand | 115 | 862_specific_main_close_interesting |
| 863 | fine - situation - features - thought - previous | 115 | 863_fine_situation_features_thought |
| 864 | worry - alright - bad - buddy - man | 115 | 864_worry_alright_bad_buddy |
| 865 | social - digital - data - users - network | 115 | 865_social_digital_data_users |
| 866 | plans - excited - secure - cryptocurrency - fees | 115 | 866_plans_excited_secure_cryptocurrency |
| 867 | media - social - available - good - early | 115 | 867_media_social_available_good |
| 868 | real - profit - investors - daily - investment | 115 | 868_real_profit_investors_daily |
| 869 | rest - buddy - mate - friend - thanks | 115 | 869_rest_buddy_mate_friend |
| 870 | issue - feature - normal - phone - difference | 115 | 870_issue_feature_normal_phone |
| 871 | alright - wonder - man - questions - words | 114 | 871_alright_wonder_man_questions |
| 872 | swap - binance - plan - announcement - related | 114 | 872_swap_binance_plan_announcement |
| 873 | head - sounds - damn - fine - note | 114 | 873_head_sounds_damn_fine |
| 874 | specific - longer - video - watch - simple | 114 | 874_specific_longer_video_watch |
| 875 | global - play - update - version - higher | 114 | 875_global_play_update_version |
| 876 | huge - cryptocurrency - address - profits - website | 114 | 876_huge_cryptocurrency_address_profits |
| 877 | network - night - bad - problem - issues | 113 | 877_network_night_bad_problem |
| 878 | crazy - smart - brother - list - man | 113 | 878_crazy_smart_brother_list |
| 879 | major - single - multiple - information - service | 113 | 879_major_single_multiple_information |
| 880 | product - date - tomorrow - announcement - version | 113 | 880_product_date_tomorrow_announcement |
| 881 | feature - try - better - lots - phone | 112 | 881_feature_try_better_lots |
| 882 | shit - crazy - wonder - smart - kind | 112 | 882_shit_crazy_wonder_smart |
| 883 | event - link - update - protocol - hours | 112 | 883_event_link_update_protocol |
| 884 | focus - building - reach - type - search | 112 | 884_focus_building_reach_type |
| 885 | version - latest - old - difference - previous | 112 | 885_version_latest_old_difference |
| 886 | multiple - - - - | 112 | 886_multiple___ |
| 887 | trading - crypto - volume - period - exchange | 112 | 887_trading_crypto_volume_period |
| 888 | spot - guy - nice - listing - live | 112 | 888_spot_guy_nice_listing |
| 889 | case - follow - article - matter - course | 111 | 889_case_follow_article_matter |
| 890 | projects - project - shit - dev - legit | 111 | 890_projects_project_shit_dev |
| 891 | claim - tokens - token - listing - utc | 111 | 891_claim_tokens_token_listing |
| 892 | thanks - information - info - mate - man | 111 | 892_thanks_information_info_mate |
| 893 | video - ethereum - sure - days - people | 111 | 893_video_ethereum_sure_days |
| 894 | tokens - token - old - contract - address | 111 | 894_tokens_token_old_contract |
| 895 | issues - fake - hard - thing - extra | 111 | 895_issues_fake_hard_thing |
| 896 | wish - chance - reward - sign - risk | 111 | 896_wish_chance_reward_sign |
| 897 | face - kind - things - sir - person | 111 | 897_face_kind_things_sir |
| 898 | payment - time - users - - | 111 | 898_payment_time_users_ |
| 899 | family - welcome - dear - love - brother | 110 | 899_family_welcome_dear_love |
| 900 | transaction - note - date - payment - account | 110 | 900_transaction_note_date_payment |
| 901 | ai - technology - problems - blockchain - smart | 110 | 901_ai_technology_problems_blockchain |
| 902 | country - difference - idea - kinda - shit | 110 | 902_country_difference_idea_kinda |
| 903 | post - thanks - work - group - telegram | 110 | 903_post_thanks_work_group |
| 904 | man - single - real - interested - friend | 109 | 904_man_single_real_interested |
| 905 | solution - features - benefits - mind - link | 109 | 905_solution_features_benefits_mind |
| 906 | eth - fees - fee - high - ethereum | 109 | 906_eth_fees_fee_high |
| 907 | global - - - - | 109 | 907_global___ |
| 908 | moment - little - trust - smart - features | 109 | 908_moment_little_trust_smart |
| 909 | morning - research - tbh - close - shit | 109 | 909_morning_research_tbh_close |
| 910 | idk - bit - case - brother - announcement | 109 | 910_idk_bit_case_brother |
| 911 | shit - man - country - dude - success | 109 | 911_shit_man_country_dude |
| 912 | service - contact - support - care - matter | 108 | 912_service_contact_support_care |
| 913 | code - clear - option - look - main | 108 | 913_code_clear_option_look |
| 914 | message - channel - available - today - updates | 108 | 914_message_channel_available_today |
| 915 | chart - trade - position - market - buy | 108 | 915_chart_trade_position_market |
| 916 | members - use - day - man - hours | 108 | 916_members_use_day_man |
| 917 | rate - worth - plan - investment - link | 108 | 917_rate_worth_plan_investment |
| 918 | recent - family - transactions - times - amazing | 108 | 918_recent_family_transactions_times |
| 919 | product - alright - dude - protocol - bad | 108 | 919_product_alright_dude_protocol |
| 920 | moving - positive - strong - assets - buying | 108 | 920_moving_positive_strong_assets |
| 921 | mind - care - opportunity - successful - change | 108 | 921_mind_care_opportunity_successful |
| 922 | positive - things - thinking - bad - point | 108 | 922_positive_things_thinking_bad |
| 923 | chat - fun - normal - interested - single | 107 | 923_chat_fun_normal_interested |
| 924 | version - latest - update - users - feature | 107 | 924_version_latest_update_users |
| 925 | experience - glad - brother - true - kind | 107 | 925_experience_glad_brother_true |
| 926 | fix - wallet - connect - issue - issues | 107 | 926_fix_wallet_connect_issue |
| 927 | code - rate - country - btc - message | 107 | 927_code_rate_country_btc |
| 928 | happy - morning - hope - mate - nice | 107 | 928_happy_morning_hope_mate |
| 929 | word - action - words - normal - crazy | 107 | 929_word_action_words_normal |
| 930 | times - similar - couple - likely - idea | 107 | 930_times_similar_couple_likely |
| 931 | fun - chat - single - real - dude | 107 | 931_fun_chat_single_real |
| 932 | new - couple - old - wonder - known | 107 | 932_new_couple_old_wonder |
| 933 | situation - issues - fact - trust - funds | 107 | 933_situation_issues_fact_trust |
| 934 | public - link - smart - date - nice | 107 | 934_public_link_smart_date |
| 935 | profits - investment - - - | 107 | 935_profits_investment__ |
| 936 | safe - country - payment - lot - message | 107 | 936_safe_country_payment_lot |
| 937 | awesome - minimum - true - deposit - profit | 107 | 937_awesome_minimum_true_deposit |
| 938 | kinda - tho - hand - person - kind | 106 | 938_kinda_tho_hand_person |
| 939 | worth - higher - product - price - web | 106 | 939_worth_higher_product_price |
| 940 | code - brother - rewards - web - similar | 106 | 940_code_brother_rewards_web |
| 941 | running - help - sorry - building - kind | 106 | 941_running_help_sorry_building |
| 942 | game - account - main - single - guys | 106 | 942_game_account_main_single |
| 943 | pepe - biggest - fair - massive - launch | 106 | 943_pepe_biggest_fair_massive |
| 944 | dude - situation - country - old - friend | 106 | 944_dude_situation_country_old |
| 945 | pool - period - utc - event - spot | 106 | 945_pool_period_utc_event |
| 946 | date - total - type - minutes - brother | 106 | 946_date_total_type_minutes |
| 947 | specific - option - kind - example - sir | 106 | 947_specific_option_kind_example |
| 948 | volume - feature - phone - low - type | 106 | 948_volume_feature_phone_low |
| 949 | brother - dude - situation - thanks - mind | 106 | 949_brother_dude_situation_thanks |
| 950 | showing - normal - case - times - transaction | 105 | 950_showing_normal_case_times |
| 951 | lots - strong - care - perfect - plans | 105 | 951_lots_strong_care_perfect |
| 952 | night - version - video - couple - brother | 105 | 952_night_version_video_couple |
| 953 | late - sense - years - reply - early | 105 | 953_late_sense_years_reply |
| 954 | imo - volume - increase - clear - major | 105 | 954_imo_volume_increase_clear |
| 955 | ama - reward - rewards - head - special | 105 | 955_ama_reward_rewards_head |
| 956 | previous - close - short - break - option | 105 | 956_previous_close_short_break |
| 957 | number - group - feature - reply - friends | 104 | 957_number_group_feature_reply |
| 958 | word - search - type - channel - profit | 104 | 958_word_search_type_channel |
| 959 | fun - - - - | 104 | 959_fun___ |
| 960 | tokens - token - participate - hand - admin | 104 | 960_tokens_token_participate_hand |
| 961 | admin - dm - doubt - guys - email | 104 | 961_admin_dm_doubt_guys |
| 962 | spot - early - listing - wrong - half | 104 | 962_spot_early_listing_wrong |
| 963 | utc - link - protocol - fine - list | 104 | 963_utc_link_protocol_fine |
| 964 | online - kinda - fine - hour - problems | 103 | 964_online_kinda_fine_hour |
| 965 | increase - day - bit - yesterday - value | 103 | 965_increase_day_bit_yesterday |
| 966 | rate - deal - volume - possible - smart | 103 | 966_rate_deal_volume_possible |
| 967 | luck - chance - couple - higher - power | 103 | 967_luck_chance_couple_higher |
| 968 | word - term - sell - specific - hold | 103 | 968_word_term_sell_specific |
| 969 | country - friend - nice - old - interesting | 102 | 969_country_friend_nice_old |
| 970 | cause - likely - reason - previous - problem | 102 | 970_cause_likely_reason_previous |
| 971 | high - active - page - - | 102 | 971_high_active_page_ |
| 972 | utc - new - pool - period - btc | 102 | 972_utc_new_pool_period |
| 973 | current - high - similar - likely - sense | 102 | 973_current_high_similar_likely |
| 974 | help - sir - question - morning - friend | 101 | 974_help_sir_question_morning |
| 975 | data - connect - solution - issue - phone | 101 | 975_data_connect_solution_issue |
| 976 | late - likely - coins - single - playing | 101 | 976_late_likely_coins_single |
| 977 | play - playing - game - games - btw | 101 | 977_play_playing_game_games |
| 978 | phone - guide - update - fix - tho | 101 | 978_phone_guide_update_fix |
| 979 | earlier - issue - hour - kind - position | 101 | 979_earlier_issue_hour_kind |
| 980 | happy - man - thanks - today - platform | 101 | 980_happy_man_thanks_today |
| 981 | - - - - | 100 | 981____ |
| 982 | complete - friend - cool - specific - wrong | 100 | 982_complete_friend_cool_specific |
| 983 | utc - listing - pm - feature - trading | 100 | 983_utc_listing_pm_feature |
| 984 | case - happy - use - day - wrong | 100 | 984_case_happy_use_day |
| 985 | business - normal - similar - assets - sense | 100 | 985_business_normal_similar_assets |
| 986 | video - account - youtube - stuff - earlier | 100 | 986_video_account_youtube_stuff |
| 987 | rate - plans - place - amazing - different | 100 | 987_rate_plans_place_amazing |
| 988 | dude - country - cool - specific - sense | 100 | 988_dude_country_cool_specific |
| 989 | option - dear - app - play - links | 100 | 989_option_dear_app_play |
| 990 | payment - pay - problems - fix - app | 100 | 990_payment_pay_problems_fix |
| 991 | details - ethereum - bot - contract - coin | 100 | 991_details_ethereum_bot_contract |
| 992 | group - admin - biggest - half - chat | 100 | 992_group_admin_biggest_half |
| 993 | page - message - post - ready - pay | 100 | 993_page_message_post_ready |
| 994 | night - country - fake - god - dude | 100 | 994_night_country_fake_god |
| 995 | product - blockchain - protocol - main - development | 99 | 995_product_blockchain_protocol_main |
| 996 | position - open - short - moment - long | 99 | 996_position_open_short_moment |
| 997 | user - new - - - | 99 | 997_user_new__ |
| 998 | benefits - step - level - specific - special | 99 | 998_benefits_step_level_specific |
| 999 | spot - order - special - face - new | 99 | 999_spot_order_special_face |
| 1000 | withdrawal - investment - platform - link - | 99 | 1000_withdrawal_investment_platform_link |
| 1001 | mind - profits - work - wallet - join | 99 | 1001_mind_profits_work_wallet |
| 1002 | services - today - business - feature - product | 99 | 1002_services_today_business_feature |
| 1003 | business - mind - tho - kinda - type | 99 | 1003_business_mind_tho_kinda |
| 1004 | chart - pepe - protocol - chain - happy | 99 | 1004_chart_pepe_protocol_chain |
| 1005 | watch - free - time - number - possible | 99 | 1005_watch_free_time_number |
| 1006 | capital - coins - transaction - fee - claim | 99 | 1006_capital_coins_transaction_fee |
| 1007 | rewards - reward - months - similar - period | 99 | 1007_rewards_reward_months_similar |
| 1008 | sir - comment - email - issue - update | 99 | 1008_sir_comment_email_issue |
| 1009 | shows - latest - report - high - profit | 99 | 1009_shows_latest_report_high |
| 1010 | friend - friends - buddy - site - mate | 99 | 1010_friend_friends_buddy_site |
| 1011 | months - word - search - chance - private | 99 | 1011_months_word_search_chance |
| 1012 | search - media - hours - service - channel | 99 | 1012_search_media_hours_service |
| 1013 | page - small - active - media - social | 99 | 1013_page_small_active_media |
| 1014 | single - register - legit - works - world | 99 | 1014_single_register_legit_works |
| 1015 | media - social - limited - user - | 98 | 1015_media_social_limited_user |
| 1016 | brother - special - reply - open - biggest | 98 | 1016_brother_special_reply_open |
| 1017 | opinion - example - ways - fun - guy | 98 | 1017_opinion_example_ways_fun |
| 1018 | public - small - old - chain - contact | 98 | 1018_public_small_old_chain |
| 1019 | rules - person - crazy - job - mind | 98 | 1019_rules_person_crazy_job |
| 1020 | data - smart - solution - hand - network | 98 | 1020_data_smart_solution_hand |
| 1021 | different - send - - - | 98 | 1021_different_send__ |
| 1022 | hit - late - close - situation - years | 98 | 1022_hit_late_close_situation |
| 1023 | chain - updates - app - benefits - users | 98 | 1023_chain_updates_app_benefits |
| 1024 | welcome - bit - announcement - bot - shit | 98 | 1024_welcome_bit_announcement_bot |
| 1025 | damn - dude - close - weeks - dev | 97 | 1025_damn_dude_close_weeks |
| 1026 | shit - sorry - word - attention - dude | 97 | 1026_shit_sorry_word_attention |
| 1027 | action - possible - - - | 97 | 1027_action_possible__ |
| 1028 | interesting - believe - community - biggest - idea | 97 | 1028_interesting_believe_community_biggest |
| 1029 | register - guide - site - face - easy | 97 | 1029_register_guide_site_face |
| 1030 | transaction - details - mate - look - block | 97 | 1030_transaction_details_mate_look |
| 1031 | volume - biggest - price - group - trading | 97 | 1031_volume_biggest_price_group |
| 1032 | fast - minutes - quick - answer - wonder | 97 | 1032_fast_minutes_quick_answer |
| 1033 | massive - services - fees - level - high | 97 | 1033_massive_services_fees_level |
| 1034 | crazy - tbh - mind - bad - lol | 97 | 1034_crazy_tbh_mind_bad |
| 1035 | live - video - similar - couple - protocol | 97 | 1035_live_video_similar_couple |
| 1036 | release - phone - updates - user - version | 97 | 1036_release_phone_updates_user |
| 1037 | recent - second - real - holding - successful | 96 | 1037_recent_second_real_holding |
| 1038 | plans - secure - financial - cryptocurrency - safe | 96 | 1038_plans_secure_financial_cryptocurrency |
| 1039 | listing - admin - contact - form - person | 96 | 1039_listing_admin_contact_form |
| 1040 | worth - tho - mate - cool - sure | 96 | 1040_worth_tho_mate_cool |
| 1041 | account - - - - | 96 | 1041_account___ |
| 1042 | upcoming - excited - cool - opinion - situation | 96 | 1042_upcoming_excited_cool_opinion |
| 1043 | search - opinion - trust - term - fact | 96 | 1043_search_opinion_trust_term |
| 1044 | specific - low - code - position - close | 96 | 1044_specific_low_code_position |
| 1045 | nice - little - ready - daily - way | 96 | 1045_nice_little_ready_daily |
| 1046 | massive - family - month - day - happy | 96 | 1046_massive_family_month_day |
| 1047 | phone - solution - fix - issue - normal | 96 | 1047_phone_solution_fix_issue |
| 1048 | interested - biggest - cause - shit - chance | 95 | 1048_interested_biggest_cause_shit |
| 1049 | close - shows - future - positive - spot | 95 | 1049_close_shows_future_positive |
| 1050 | volume - chart - liquidity - low - ethereum | 95 | 1050_volume_chart_liquidity_low |
| 1051 | selling - contact - multiple - personal - access | 95 | 1051_selling_contact_multiple_personal |
| 1052 | btc - coins - coin - previous - value | 95 | 1052_btc_coins_coin_previous |
| 1053 | eth - transaction - pool - yesterday - funds | 95 | 1053_eth_transaction_pool_yesterday |
| 1054 | game - announcement - discord - specific - visit | 95 | 1054_game_announcement_discord_specific |
| 1055 | details - matter - mind - similar - specific | 95 | 1055_details_matter_mind_similar |
| 1056 | eth - ethereum - difference - network - chain | 95 | 1056_eth_ethereum_difference_network |
| 1057 | fee - benefits - power - product - lots | 95 | 1057_fee_benefits_power_product |
| 1058 | details - bot - contract - address - exciting | 95 | 1058_details_bot_contract_address |
| 1059 | ethereum - major - benefits - number - little | 94 | 1059_ethereum_major_benefits_number |
| 1060 | half - interested - chat - reply - face | 94 | 1060_half_interested_chat_reply |
| 1061 | country - dude - specific - group - situation | 94 | 1061_country_dude_specific_group |
| 1062 | payment - option - development - digital - main | 94 | 1062_payment_option_development_digital |
| 1063 | huge - big - wonder - small - biggest | 94 | 1063_huge_big_wonder_small |
| 1064 | god - day - good - morning - thanks | 94 | 1064_god_day_good_morning |
| 1065 | ai - wallet - app - blockchain - early | 94 | 1065_ai_wallet_app_blockchain |
| 1066 | problems - hand - idk - research - opinion | 94 | 1066_problems_hand_idk_research |
| 1067 | option - sense - assets - example - claim | 94 | 1067_option_sense_assets_example |
| 1068 | huge - financial - success - fees - live | 94 | 1068_huge_financial_success_fees |
| 1069 | fair - tomorrow - months - true - launch | 93 | 1069_fair_tomorrow_months_true |
| 1070 | version - game - excited - phone - games | 93 | 1070_version_game_excited_phone |
| 1071 | super - hand - cool - mind - dude | 93 | 1071_super_hand_cool_mind |
| 1072 | lol - years - problems - reply - buddy | 93 | 1072_lol_years_problems_reply |
| 1073 | financial - opinion - situation - sorry - wonder | 93 | 1073_financial_opinion_situation_sorry |
| 1074 | bit - awesome - job - special - hand | 93 | 1074_bit_awesome_job_special |
| 1075 | network - buddy - features - building - web | 93 | 1075_network_buddy_features_building |
| 1076 | ways - sorry - ppl - bit - course | 93 | 1076_ways_sorry_ppl_bit |
| 1077 | ecosystem - projects - future - special - protocol | 93 | 1077_ecosystem_projects_future_special |
| 1078 | page - big - great - day - benefits | 92 | 1078_page_big_great_day |
| 1079 | questions - smart - specific - free - bot | 92 | 1079_questions_smart_specific_free |
| 1080 | cryptocurrency - big - contact - help - investment | 92 | 1080_cryptocurrency_big_contact_help |
| 1081 | page - message - great - available - limited | 92 | 1081_page_message_great_available |
| 1082 | head - reply - crazy - dev - moment | 92 | 1082_head_reply_crazy_dev |
| 1083 | hand - hold - playing - super - haha | 92 | 1083_hand_hold_playing_super |
| 1084 | issue - max - phone - normal - issues | 92 | 1084_issue_max_phone_normal |
| 1085 | fix - dev - solution - issue - fine | 92 | 1085_fix_dev_solution_issue |
| 1086 | available - info - awesome - thanks - great | 92 | 1086_available_info_awesome_thanks |
| 1087 | active - available - today - good - | 92 | 1087_active_available_today_good |
| 1088 | correct - similar - idea - issue - certain | 92 | 1088_correct_similar_idea_issue |
| 1089 | term - holding - capital - long - investment | 92 | 1089_term_holding_capital_long |
| 1090 | fact - main - certain - super - place | 91 | 1090_fact_main_certain_super |
| 1091 | rules - - - - | 91 | 1091_rules___ |
| 1092 | higher - current - wait - ask - price | 91 | 1092_higher_current_wait_ask |
| 1093 | kinda - situation - reason - yesterday - earlier | 91 | 1093_kinda_situation_reason_yesterday |
| 1094 | build - latest - official - old - issues | 91 | 1094_build_latest_official_old |
| 1095 | sorry - bit - head - sure - article | 91 | 1095_sorry_bit_head_sure |
| 1096 | family - digital - payment - address - access | 91 | 1096_family_digital_payment_address |
| 1097 | head - wonder - job - certain - bad | 91 | 1097_head_wonder_job_certain |
| 1098 | order - group - action - links - user | 90 | 1098_order_group_action_links |
| 1099 | fine - bad - awesome - okay - cool | 90 | 1099_fine_bad_awesome_okay |
| 1100 | web - strong - care - user - security | 90 | 1100_web_strong_care_user |
| 1101 | user - sir - hand - multiple - set | 90 | 1101_user_sir_hand_multiple |
| 1102 | case - plan - global - important - family | 90 | 1102_case_plan_global_important |
| 1103 | moment - games - couple - sense - sorry | 90 | 1103_moment_games_couple_sense |
| 1104 | course - guide - case - ready - access | 90 | 1104_course_guide_case_ready |
| 1105 | pepe - fun - marketing - time - wallet | 90 | 1105_pepe_fun_marketing_time |
| 1106 | public - small - old - chain - contact | 90 | 1106_public_small_old_chain |
| 1107 | nice - day - great - friend - guys | 89 | 1107_nice_day_great_friend |
| 1108 | network - app - code - blockchain - update | 89 | 1108_network_app_code_blockchain |
| 1109 | successful - fake - different - holding - near | 89 | 1109_successful_fake_different_holding |
| 1110 | rest - hope - legit - fake - face | 89 | 1110_rest_hope_legit_fake |
| 1111 | win - risk - level - official - daily | 89 | 1111_win_risk_level_official |
| 1112 | team - reply - answer - details - situation | 89 | 1112_team_reply_answer_details |
| 1113 | interesting - night - break - shows - cool | 89 | 1113_interesting_night_break_shows |
| 1114 | block - nice - fine - sense - opinion | 89 | 1114_block_nice_fine_sense |
| 1115 | man - pm - cool - early - head | 89 | 1115_man_pm_cool_early |
| 1116 | site - issues - website - problems - fine | 89 | 1116_site_issues_website_problems |
| 1117 | situation - form - action - okay - reason | 89 | 1117_situation_form_action_okay |
| 1118 | man - mate - event - family - period | 89 | 1118_man_mate_event_family |
| 1119 | doubt - clear - ways - mind - course | 89 | 1119_doubt_clear_ways_mind |
| 1120 | recent - report - nice - - | 89 | 1120_recent_report_nice_ |
| 1121 | ones - head - care - run - game | 89 | 1121_ones_head_care_run |
| 1122 | works - work - tho - thing - interesting | 88 | 1122_works_work_tho_thing |
| 1123 | projects - market - release - project - product | 88 | 1123_projects_market_release_project |
| 1124 | site - video - available - day - new | 88 | 1124_site_video_available_day |
| 1125 | fees - transaction - transactions - development - event | 88 | 1125_fees_transaction_transactions_development |
| 1126 | step - post - personal - event - rules | 88 | 1126_step_post_personal_event |
| 1127 | sense - half - feature - opinion - control | 88 | 1127_sense_half_feature_opinion |
| 1128 | massive - world - fees - safe - active | 88 | 1128_massive_world_fees_safe |
| 1129 | guide - correct - option - solution - weeks | 88 | 1129_guide_correct_option_solution |
| 1130 | data - hand - mentioned - wrong - problems | 88 | 1130_data_hand_mentioned_wrong |
| 1131 | unique - global - doubt - successful - investing | 88 | 1131_unique_global_doubt_successful |
| 1132 | means - register - profit - money - limited | 88 | 1132_means_register_profit_money |
| 1133 | fun - wan - dm - real - single | 88 | 1133_fun_wan_dm_real |
| 1134 | fine - okay - buddy - dude - thanks | 88 | 1134_fine_okay_buddy_dude |
| 1135 | opportunity - deposit - investors - bitcoin - hours | 87 | 1135_opportunity_deposit_investors_bitcoin |
| 1136 | country - guy - welcome - bit - feel | 87 | 1136_country_guy_welcome_bit |
| 1137 | mind - special - guy - brother - bro | 87 | 1137_mind_special_guy_brother |
| 1138 | difference - mind - life - experience - world | 87 | 1138_difference_mind_life_experience |
| 1139 | active - - - - | 87 | 1139_active___ |
| 1140 | comment - reply - specific - point - clear | 87 | 1140_comment_reply_specific_point |
| 1141 | project - projects - idea - certain - man | 87 | 1141_project_projects_idea_certain |
| 1142 | ethereum - low - exchange - trade - half | 87 | 1142_ethereum_low_exchange_trade |
| 1143 | feature - swap - excited - step - digital | 87 | 1143_feature_swap_excited_step |
| 1144 | potential - lot - huge - cool - likely | 87 | 1144_potential_lot_huge_cool |
| 1145 | hand - type - channel - link - telegram | 87 | 1145_hand_type_channel_link |
| 1146 | friend - bit - situation - difference - dear | 87 | 1146_friend_bit_situation_difference |
| 1147 | half - ready - awesome - old - buddy | 87 | 1147_half_ready_awesome_old |
| 1148 | exchange - coin - games - features - ecosystem | 87 | 1148_exchange_coin_games_features |
| 1149 | similar - user - update - note - article | 86 | 1149_similar_user_update_note |
| 1150 | list - known - site - hand - high | 86 | 1150_list_known_site_hand |
| 1151 | volume - price - free - trading - month | 86 | 1151_volume_price_free_trading |
| 1152 | mind - head - country - clear - fact | 86 | 1152_mind_head_country_clear |
| 1153 | technology - holders - network - little - couple | 86 | 1153_technology_holders_network_little |
| 1154 | ecosystem - volume - recent - news - value | 86 | 1154_ecosystem_volume_recent_news |
| 1155 | exchange - account - sir - step - funds | 86 | 1155_exchange_account_sir_step |
| 1156 | support - thanks - kind - man - guys | 86 | 1156_support_thanks_kind_man |
| 1157 | group - chat - talk - main - problem | 86 | 1157_group_chat_talk_main |
| 1158 | early - come - form - moment - chance | 86 | 1158_early_come_form_moment |
| 1159 | news - crypto - investors - cryptocurrency - crazy | 86 | 1159_news_crypto_investors_cryptocurrency |
| 1160 | spot - fees - global - open - users | 86 | 1160_spot_fees_global_open |
| 1161 | product - important - little - questions - links | 86 | 1161_product_important_little_questions |
| 1162 | certain - shit - hand - research - space | 85 | 1162_certain_shit_hand_research |
| 1163 | claim - ecosystem - transactions - tokens - today | 85 | 1163_claim_ecosystem_transactions_tokens |
| 1164 | dude - ok - guy - bro - word | 85 | 1164_dude_ok_guy_bro |
| 1165 | welcome - ama - form - issues - global | 85 | 1165_welcome_ama_form_issues |
| 1166 | legit - site - guy - related - questions | 85 | 1166_legit_site_guy_related |
| 1167 | utc - pm - tomorrow - time - morning | 85 | 1167_utc_pm_tomorrow_time |
| 1168 | tokens - period - token - specific - half | 85 | 1168_tokens_period_token_specific |
| 1169 | live - game - games - idea - interesting | 85 | 1169_live_game_games_idea |
| 1170 | global - words - kindly - users - alot | 85 | 1170_global_words_kindly_users |
| 1171 | bit - couple - problems - game - step | 85 | 1171_bit_couple_problems_game |
| 1172 | news - knows - clear - true - likely | 84 | 1172_news_knows_clear_true |
| 1173 | release - half - early - likely - place | 84 | 1173_release_half_early_likely |
| 1174 | change - difference - stuff - positive - haha | 84 | 1174_change_difference_stuff_positive |
| 1175 | previous - step - luck - months - similar | 84 | 1175_previous_step_luck_months |
| 1176 | hope - huge - buddy - sorry - quick | 84 | 1176_hope_huge_buddy_sorry |
| 1177 | money - dude - shit - brother - couple | 84 | 1177_money_dude_shit_brother |
| 1178 | address - number - btc - little - plans | 84 | 1178_address_number_btc_little |
| 1179 | release - date - near - specific - opinion | 84 | 1179_release_date_near_specific |
| 1180 | head - period - near - situation - months | 84 | 1180_head_period_near_situation |
| 1181 | opinion - low - couple - certain - development | 84 | 1181_opinion_low_couple_certain |
| 1182 | term - short - huge - man - research | 84 | 1182_term_short_huge_man |
| 1183 | period - benefits - utc - listing - pm | 84 | 1183_period_benefits_utc_listing |
| 1184 | control - country - late - open - secure | 83 | 1184_control_country_late_open |
| 1185 | minimum - deposit - daily - withdrawal - site | 83 | 1185_minimum_deposit_daily_withdrawal |
| 1186 | article - visit - dear - information - better | 83 | 1186_article_visit_dear_information |
| 1187 | brother - person - awesome - glad - luck | 83 | 1187_brother_person_awesome_glad |
| 1188 | ecosystem - interesting - chain - guess - public | 83 | 1188_ecosystem_interesting_chain_guess |
| 1189 | services - post - information - message - channel | 83 | 1189_services_post_information_message |
| 1190 | weeks - brother - hour - risk - long | 83 | 1190_weeks_brother_hour_risk |
| 1191 | issue - data - saw - version - projects | 83 | 1191_issue_data_saw_version |
| 1192 | reply - bit - man - thing - fun | 83 | 1192_reply_bit_man_thing |
| 1193 | project - price - projects - situation - development | 83 | 1193_project_price_projects_situation |
| 1194 | position - chart - holders - funds - price | 83 | 1194_position_chart_holders_funds |
| 1195 | transaction - fee - fees - transactions - network | 83 | 1195_transaction_fee_fees_transactions |
| 1196 | positive - love - kinda - man - cool | 83 | 1196_positive_love_kinda_man |
| 1197 | early - sounds - type - moment - saying | 83 | 1197_early_sounds_type_moment |
| 1198 | tokens - token - total - problems - drop | 83 | 1198_tokens_token_total_problems |
| 1199 | fact - type - word - search - funds | 83 | 1199_fact_type_word_search |
| 1200 | guess - game - bad - opinion - fun | 82 | 1200_guess_game_bad_opinion |
| 1201 | interesting - attention - major - protocol - value | 82 | 1201_interesting_attention_major_protocol |
| 1202 | clear - point - starting - cause - night | 82 | 1202_clear_point_starting_cause |
| 1203 | transactions - transaction - single - number - similar | 82 | 1203_transactions_transaction_single_number |
| 1204 | night - tomorrow - good - alright - nice | 82 | 1204_night_tomorrow_good_alright |
| 1205 | key - fees - process - cryptocurrency - experience | 82 | 1205_key_fees_process_cryptocurrency |
| 1206 | list - guy - couple - problems - guess | 82 | 1206_list_guy_couple_problems |
| 1207 | upcoming - key - btc - search - channel | 82 | 1207_upcoming_key_btc_search |
| 1208 | ethereum - contract - correct - btc - wrong | 82 | 1208_ethereum_contract_correct_btc |
| 1209 | power - trade - right - profit - years | 82 | 1209_power_trade_right_profit |
| 1210 | nft - point - friend - power - half | 82 | 1210_nft_point_friend_power |
| 1211 | online - money - life - platform - great | 82 | 1211_online_money_life_platform |
| 1212 | launch - multiple - huge - chain - marketing | 81 | 1212_launch_multiple_huge_chain |
| 1213 | excited - period - exciting - strong - holding | 81 | 1213_excited_period_exciting_strong |
| 1214 | words - date - specific - years - product | 81 | 1214_words_date_specific_years |
| 1215 | hand - shit - ones - wan - man | 81 | 1215_hand_shit_ones_wan |
| 1216 | token - fake - hand - tokens - site | 81 | 1216_token_fake_hand_tokens |
| 1217 | - - - - | 81 | 1217____ |
| 1218 | app - account - issues - page - version | 81 | 1218_app_account_issues_page |
| 1219 | attention - cool - open - certain - opinion | 81 | 1219_attention_cool_open_certain |
| 1220 | offer - special - email - buddy - kinda | 81 | 1220_offer_special_email_buddy |
| 1221 | minutes - live - awesome - week - latest | 81 | 1221_minutes_live_awesome_week |
| 1222 | play - games - game - fun - earning | 81 | 1222_play_games_game_fun |
| 1223 | months - focus - main - world - ta | 81 | 1223_months_focus_main_world |
| 1224 | dude - bro - guy - cool - situation | 81 | 1224_dude_bro_guy_cool |
| 1225 | end - total - sorry - check - new | 81 | 1225_end_total_sorry_check |
| 1226 | legit - fee - post - try - sure | 80 | 1226_legit_fee_post_try |
| 1227 | ready - group - - - | 80 | 1227_ready_group__ |
| 1228 | glad - positive - features - interested - questions | 80 | 1228_glad_positive_features_interested |
| 1229 | correct - dude - brother - situation - end | 80 | 1229_correct_dude_brother_situation |
| 1230 | crazy - pepe - kind - little - bit | 80 | 1230_crazy_pepe_kind_little |
| 1231 | nice - good - tho - cool - great | 80 | 1231_nice_good_tho_cool |
| 1232 | period - utc - yesterday - couple - feature | 80 | 1232_period_utc_yesterday_couple |
| 1233 | lol - shit - fine - bit - job | 80 | 1233_lol_shit_fine_bit |
| 1234 | tomorrow - risk - development - information - future | 80 | 1234_tomorrow_risk_development_information |
| 1235 | point - small - wrong - reason - course | 80 | 1235_point_small_wrong_reason |
| 1236 | page - small - active - media - social | 80 | 1236_page_small_active_media |
| 1237 | buddy - mate - best - thanks - luck | 80 | 1237_buddy_mate_best_thanks |
| 1238 | minimum - deposit - daily - reward - site | 80 | 1238_minimum_deposit_daily_reward |
| 1239 | exchange - announcement - listing - specific - exciting | 80 | 1239_exchange_announcement_listing_specific |
| 1240 | situation - guy - positive - point - things | 80 | 1240_situation_guy_positive_point |
| 1241 | interesting - type - dude - sense - opinion | 80 | 1241_interesting_type_dude_sense |
| 1242 | capital - holding - plan - opportunity - best | 79 | 1242_capital_holding_plan_opportunity |
| 1243 | opinion - hand - past - upcoming - features | 79 | 1243_opinion_hand_past_upcoming |
| 1244 | day - free - tho - wrong - friend | 79 | 1244_day_free_tho_wrong |
| 1245 | projects - months - project - guy - research | 79 | 1245_projects_months_project_guy |
| 1246 | weeks - months - days - buddy - month | 79 | 1246_weeks_months_days_buddy |
| 1247 | pool - funds - longer - normal - product | 79 | 1247_pool_funds_longer_normal |
| 1248 | financial - services - years - experience - capital | 79 | 1248_financial_services_years_experience |
| 1249 | reward - bot - level - deposit - minimum | 79 | 1249_reward_bot_level_deposit |
| 1250 | guide - version - friend - similar - google | 79 | 1250_guide_version_friend_similar |
| 1251 | opinion - true - sense - personal - correct | 79 | 1251_opinion_true_sense_personal |
| 1252 | face - short - long - ones - dude | 79 | 1252_face_short_long_ones |
| 1253 | ones - wonder - hand - similar - digital | 79 | 1253_ones_wonder_hand_similar |
| 1254 | hours - phone - drop - pay - real | 79 | 1254_hours_phone_drop_pay |
| 1255 | nft - sir - network - chain - solution | 79 | 1255_nft_sir_network_chain |
| 1256 | luck - good - buddy - mate - cool | 79 | 1256_luck_good_buddy_mate |
| 1257 | wait - wallet - - - | 79 | 1257_wait_wallet__ |
| 1258 | level - minimum - plan - daily - fee | 79 | 1258_level_minimum_plan_daily |
| 1259 | global - crazy - country - specific - happening | 78 | 1259_global_crazy_country_specific |
| 1260 | situation - nice - old - dude - friend | 78 | 1260_situation_nice_old_dude |
| 1261 | dev - early - shit - excited - video | 78 | 1261_dev_early_shit_excited |
| 1262 | upcoming - exciting - latest - public - updates | 78 | 1262_upcoming_exciting_latest_public |
| 1263 | pool - recent - near - morning - open | 78 | 1263_pool_recent_near_morning |
| 1264 | withdrawal - term - short - plan - fee | 78 | 1264_withdrawal_term_short_plan |
| 1265 | country - face - wonder - brother - control | 78 | 1265_country_face_wonder_brother |
| 1266 | report - reward - article - ecosystem - security | 78 | 1266_report_reward_article_ecosystem |
| 1267 | price - half - deal - previous - fact | 78 | 1267_price_half_deal_previous |
| 1268 | late - luck - bit - sense - opinion | 78 | 1268_late_luck_bit_sense |
| 1269 | control - transactions - previous - coin - past | 78 | 1269_control_transactions_previous_coin |
| 1270 | message - user - - - | 78 | 1270_message_user__ |
| 1271 | glad - place - head - comment - weeks | 78 | 1271_glad_place_head_comment |
| 1272 | bad - fact - stuff - fine - things | 78 | 1272_bad_fact_stuff_fine |
| 1273 | opinion - clear - question - sense - imagine | 78 | 1273_opinion_clear_question_sense |
| 1274 | interested - pm - little - stuff - ones | 78 | 1274_interested_pm_little_stuff |
| 1275 | crypto - development - control - interesting - technology | 78 | 1275_crypto_development_control_interesting |
| 1276 | phone - idea - guide - step - brother | 77 | 1276_phone_idea_guide_step |
| 1277 | feature - trading - similar - spot - trade | 77 | 1277_feature_trading_similar_spot |
| 1278 | fact - opinion - people - fun - comment | 77 | 1278_fact_opinion_people_fun |
| 1279 | reward - rewards - public - users - specific | 77 | 1279_reward_rewards_public_users |
| 1280 | specific - product - dev - wonder - word | 77 | 1280_specific_product_dev_wonder |
| 1281 | cool - real - kinda - interesting - run | 77 | 1281_cool_real_kinda_interesting |
| 1282 | previous - global - early - massive - launch | 77 | 1282_previous_global_early_massive |
| 1283 | comment - product - exciting - community - form | 77 | 1283_comment_product_exciting_community |
| 1284 | small - big - dude - biggest - night | 77 | 1284_small_big_dude_biggest |
| 1285 | home - online - earning - money - step | 76 | 1285_home_online_earning_money |
| 1286 | hit - hand - brother - stuff - web | 76 | 1286_hit_hand_brother_stuff |
| 1287 | phone - luck - shows - problems - waiting | 76 | 1287_phone_luck_shows_problems |
| 1288 | btc - year - bitcoin - cause - months | 76 | 1288_btc_year_bitcoin_cause |
| 1289 | brother - tbh - rest - similar - opinion | 76 | 1289_brother_tbh_rest_similar |
| 1290 | minutes - job - massive - message - case | 76 | 1290_minutes_job_massive_message |
| 1291 | contract - address - ethereum - eth - correct | 76 | 1291_contract_address_ethereum_eth |
| 1292 | control - sense - things - words - dude | 76 | 1292_control_sense_things_words |
| 1293 | market - coins - hope - deal - situation | 75 | 1293_market_coins_hope_deal |
| 1294 | sorry - wrong - clear - face - similar | 75 | 1294_sorry_wrong_clear_face |
| 1295 | team - price - situation - control - clear | 75 | 1295_team_price_situation_control |
| 1296 | close - chart - period - open - specific | 75 | 1296_close_chart_period_open |
| 1297 | feature - announcement - recent - rewards - stuff | 75 | 1297_feature_announcement_recent_rewards |
| 1298 | playing - super - technology - blockchain - online | 75 | 1298_playing_super_technology_blockchain |
| 1299 | head - exciting - pool - ecosystem - future | 75 | 1299_head_exciting_pool_ecosystem |
| 1300 | claim - holders - little - specific - announcement | 75 | 1300_claim_holders_little_specific |
| 1301 | power - old - couple - chance - ways | 75 | 1301_power_old_couple_chance |
| 1302 | available - info - difference - bit - chart | 75 | 1302_available_info_difference_bit |
| 1303 | phone - number - code - email - luck | 75 | 1303_phone_number_code_email |
| 1304 | face - dude - kind - guy - sense | 75 | 1304_face_dude_kind_guy |
| 1305 | email - version - available - data - phone | 75 | 1305_email_version_available_data |
| 1306 | fact - job - crazy - wants - shit | 75 | 1306_fact_job_crazy_wants |
| 1307 | protocol - data - control - couple - smart | 74 | 1307_protocol_data_control_couple |
| 1308 | members - group - contact - available - active | 74 | 1308_members_group_contact_available |
| 1309 | past - yesterday - chain - short - major | 74 | 1309_past_yesterday_chain_short |
| 1310 | work - potential - chance - - | 74 | 1310_work_potential_chance_ |
| 1311 | mind - amazing - nice - opportunity - great | 74 | 1311_mind_amazing_nice_opportunity |
| 1312 | type - hand - answer - correct - sorry | 74 | 1312_type_hand_answer_correct |
| 1313 | phone - number - code - issue - email | 74 | 1313_phone_number_code_issue |
| 1314 | fact - wonder - shit - late - stuff | 74 | 1314_fact_wonder_shit_late |
| 1315 | nft - chance - claim - rules - unique | 74 | 1315_nft_chance_claim_rules |
| 1316 | job - media - social - free - | 74 | 1316_job_media_social_free |
| 1317 | super - early - end - site - excited | 74 | 1317_super_early_end_site |
| 1318 | months - month - end - couple - half | 74 | 1318_months_month_end_couple |
| 1319 | withdrawal - amazing - research - mind - visit | 74 | 1319_withdrawal_amazing_research_mind |
| 1320 | discord - questions - recent - announcement - official | 74 | 1320_discord_questions_recent_announcement |
| 1321 | recent - holders - successful - bitcoin - support | 74 | 1321_recent_holders_successful_bitcoin |
| 1322 | luck - block - earlier - near - volume | 74 | 1322_luck_block_earlier_near |
| 1323 | glad - platform - happy - good - hours | 73 | 1323_glad_platform_happy_good |
| 1324 | write - great - shit - dude - couple | 73 | 1324_write_great_shit_dude |
| 1325 | alot - ones - issues - date - successful | 73 | 1325_alot_ones_issues_date |
| 1326 | media - social - limited - - | 73 | 1326_media_social_limited_ |
| 1327 | small - specific - couple - huge - minimum | 73 | 1327_small_specific_couple_huge |
| 1328 | data - type - version - problem - feature | 73 | 1328_data_type_version_problem |
| 1329 | power - super - special - shit - hand | 73 | 1329_power_super_special_shit |
| 1330 | event - listing - volume - hand - trading | 73 | 1330_event_listing_volume_hand |
| 1331 | address - wallet - mate - dm - private | 72 | 1331_address_wallet_mate_dm |
| 1332 | normal - man - night - bit - lol | 72 | 1332_normal_man_night_bit |
| 1333 | fine - alright - ok - nice - buddy | 72 | 1333_fine_alright_ok_nice |
| 1334 | man - problems - awesome - watch - similar | 72 | 1334_man_problems_awesome_watch |
| 1335 | access - telegram - free - - | 72 | 1335_access_telegram_free_ |
| 1336 | page - high - big - new - daily | 72 | 1336_page_high_big_new |
| 1337 | page - free - issues - small - users | 72 | 1337_page_free_issues_small |
| 1338 | friend - morning - course - normal - late | 72 | 1338_friend_morning_course_normal |
| 1339 | code - research - opinion - crypto - digital | 72 | 1339_code_research_opinion_crypto |
| 1340 | watch - worth - head - try - chance | 72 | 1340_watch_worth_head_try |
| 1341 | kinda - sense - transactions - normal - updates | 72 | 1341_kinda_sense_transactions_normal |
| 1342 | release - public - small - old - exchange | 72 | 1342_release_public_small_old |
| 1343 | job - experience - interested - position - years | 72 | 1343_job_experience_interested_position |
| 1344 | coin - hours - new - latest - | 72 | 1344_coin_hours_new_latest |
| 1345 | private - address - ethereum - sell - tokens | 72 | 1345_private_address_ethereum_sell |
| 1346 | links - services - marketing - project - price | 72 | 1346_links_services_marketing_project |
| 1347 | tokens - token - rest - option - total | 72 | 1347_tokens_token_rest_option |
| 1348 | minutes - likely - special - tomorrow - early | 72 | 1348_minutes_likely_special_tomorrow |
| 1349 | higher - feature - current - weeks - sense | 72 | 1349_higher_feature_current_weeks |
| 1350 | current - world - article - technology - wonder | 72 | 1350_current_world_article_technology |
| 1351 | technology - ecosystem - announcement - opinion - projects | 72 | 1351_technology_ecosystem_announcement_opinion |
| 1352 | opportunity - register - capital - best - platform | 72 | 1352_opportunity_register_capital_best |
| 1353 | private - dude - pm - brother - opinion | 72 | 1353_private_dude_pm_brother |
| 1354 | answer - place - cryptocurrency - level - people | 72 | 1354_answer_place_cryptocurrency_level |
| 1355 | feature - super - code - early - moment | 71 | 1355_feature_super_code_early |
| 1356 | late - weeks - year - situation - order | 71 | 1356_late_weeks_year_situation |
| 1357 | group - bot - reply - simple - questions | 71 | 1357_group_bot_reply_simple |
| 1358 | build - version - latest - problem - similar | 71 | 1358_build_version_latest_problem |
| 1359 | reply - certain - wrong - mind - tho | 71 | 1359_reply_certain_wrong_mind |
| 1360 | thanks - amazing - tomorrow - massive - super | 71 | 1360_thanks_amazing_tomorrow_massive |
| 1361 | chain - idea - issue - current - situation | 71 | 1361_chain_idea_issue_current |
| 1362 | face - quick - cool - fine - home | 71 | 1362_face_quick_cool_fine |
| 1363 | main - code - yesterday - single - couple | 71 | 1363_main_code_yesterday_single |
| 1364 | guide - step - bro - article - friend | 71 | 1364_guide_step_bro_article |
| 1365 | start - available - today - day - time | 71 | 1365_start_available_today_day |
| 1366 | successful - investment - company - - | 71 | 1366_successful_investment_company_ |
| 1367 | access - - - - | 71 | 1367_access___ |
| 1368 | level - pool - welcome - payment - free | 71 | 1368_level_pool_welcome_payment |
| 1369 | - - - - | 71 | 1369____ |
| 1370 | biggest - network - exchange - world - transactions | 71 | 1370_biggest_network_exchange_world |
| 1371 | launch - marketing - plans - coins - power | 71 | 1371_launch_marketing_plans_coins |
| 1372 | pepe - haha - early - dude - profits | 70 | 1372_pepe_haha_early_dude |
| 1373 | opinion - dude - head - man - ecosystem | 70 | 1373_opinion_dude_head_man |
| 1374 | search - certain - simple - action - stuff | 70 | 1374_search_certain_simple_action |
| 1375 | couple - kinda - drop - dude - friend | 70 | 1375_couple_kinda_drop_dude |
| 1376 | problems - digital - solution - strong - hit | 70 | 1376_problems_digital_solution_strong |
| 1377 | media - social - limited - morning - huge | 70 | 1377_media_social_limited_morning |
| 1378 | smart - guys - support - chain - ai | 70 | 1378_smart_guys_support_chain |
| 1379 | close - upcoming - couple - sense - wrong | 70 | 1379_close_upcoming_couple_sense |
| 1380 | information - guys - website - available - | 70 | 1380_information_guys_website_available |
| 1381 | thing - benefits - reason - people - power | 70 | 1381_thing_benefits_reason_people |
| 1382 | exciting - development - potential - cool - true | 70 | 1382_exciting_development_potential_cool |
| 1383 | phone - working - solution - turn - network | 70 | 1383_phone_working_solution_turn |
| 1384 | similar - btc - hand - attention - value | 70 | 1384_similar_btc_hand_attention |
| 1385 | fix - announcement - step - certain - opinion | 70 | 1385_fix_announcement_step_certain |
| 1386 | guy - rules - hand - dude - couple | 70 | 1386_guy_rules_hand_dude |
| 1387 | turn - wonder - thought - dude - world | 70 | 1387_turn_wonder_thought_dude |
| 1388 | brother - feature - break - weeks - matter | 70 | 1388_brother_feature_break_weeks |
| 1389 | early - job - head - man - live | 70 | 1389_early_job_head_man |
| 1390 | links - open - available - - | 70 | 1390_links_open_available_ |
| 1391 | fun - yesterday - night - home - brother | 70 | 1391_fun_yesterday_night_home |
| 1392 | money - bro - reason - buddy - crazy | 69 | 1392_money_bro_reason_buddy |
| 1393 | nice - - - - | 69 | 1393_nice___ |
| 1394 | shit - feel - home - brother - normal | 69 | 1394_shit_feel_home_brother |
| 1395 | pool - updates - glad - hand - dude | 69 | 1395_pool_updates_glad_hand |
| 1396 | interesting - opinion - short - major - yesterday | 69 | 1396_interesting_opinion_short_major |
| 1397 | country - likely - job - normal - people | 69 | 1397_country_likely_job_normal |
| 1398 | ask - channel - link - - | 69 | 1398_ask_channel_link_ |
| 1399 | fix - luck - claim - exciting - worth | 69 | 1399_fix_luck_claim_exciting |
| 1400 | words - group - company - best - good | 69 | 1400_words_group_company_best |
| 1401 | night - early - version - fix - couple | 69 | 1401_night_early_version_fix |
| 1402 | major - single - multiple - information - service | 69 | 1402_major_single_multiple_information |
| 1403 | live - weeks - link - public - page | 69 | 1403_live_weeks_link_public |
| 1404 | power - lol - half - change - bad | 69 | 1404_power_lol_half_change |
| 1405 | plans - thing - fees - extra - holding | 69 | 1405_plans_thing_fees_extra |
| 1406 | tokens - token - correct - late - claim | 69 | 1406_tokens_token_correct_late |
| 1407 | hard - - - - | 69 | 1407_hard___ |
| 1408 | single - benefits - tomorrow - term - opportunity | 69 | 1408_single_benefits_tomorrow_term |
| 1409 | important - space - dude - public - control | 69 | 1409_important_space_dude_public |
| 1410 | dude - biggest - shit - word - lol | 69 | 1410_dude_biggest_shit_word |
| 1411 | luck - mate - complete - wish - form | 69 | 1411_luck_mate_complete_wish |
| 1412 | thanks - kind - glad - form - mate | 69 | 1412_thanks_kind_glad_form |
| 1413 | user - problems - public - issues - question | 69 | 1413_user_problems_public_issues |
| 1414 | ethereum - week - time - - | 69 | 1414_ethereum_week_time_ |
| 1415 | course - reason - opinion - thing - way | 69 | 1415_course_reason_opinion_thing |
| 1416 | weeks - positive - deal - head - current | 69 | 1416_weeks_positive_deal_head |
| 1417 | quick - accounts - today - - | 69 | 1417_quick_accounts_today_ |
| 1418 | super - win - spot - year - date | 68 | 1418_super_win_spot_year |
| 1419 | business - digital - thing - years - marketing | 68 | 1419_business_digital_thing_years |
| 1420 | release - earn - place - hand - damn | 68 | 1420_release_earn_place_hand |
| 1421 | help - thanks - cool - buddy - guys | 68 | 1421_help_thanks_cool_buddy |
| 1422 | contract - friends - official - launch - website | 68 | 1422_contract_friends_official_launch |
| 1423 | job - work - mean - kinda - stuff | 68 | 1423_job_work_mean_kinda |
| 1424 | better - hope - okay - sorry - friend | 68 | 1424_better_hope_okay_sorry |
| 1425 | game - point - moment - half - form | 68 | 1425_game_point_moment_half |
| 1426 | strong - success - ecosystem - technology - option | 68 | 1426_strong_success_ecosystem_technology |
| 1427 | words - protocol - friend - secure - years | 68 | 1427_words_protocol_friend_secure |
| 1428 | rewards - reward - half - correct - small | 68 | 1428_rewards_reward_half_correct |
| 1429 | half - current - clear - rest - supply | 68 | 1429_half_current_clear_rest |
| 1430 | investors - bitcoin - investment - company - time | 68 | 1430_investors_bitcoin_investment_company |
| 1431 | problems - brother - change - mind - bro | 68 | 1431_problems_brother_change_mind |
| 1432 | coin - coins - reason - run - ecosystem | 68 | 1432_coin_coins_reason_run |
| 1433 | focus - things - stuff - life - game | 68 | 1433_focus_things_stuff_life |
| 1434 | project - projects - new - comment - stuff | 68 | 1434_project_projects_new_comment |
| 1435 | massive - huge - big - stuff - ones | 68 | 1435_massive_huge_big_stuff |
| 1436 | happy - token - group - main - idea | 68 | 1436_happy_token_group_main |
| 1437 | drop - forward - period - protocol - guide | 68 | 1437_drop_forward_period_protocol |
| 1438 | couple - sense - previous - data - assets | 68 | 1438_couple_sense_previous_data |
| 1439 | kinda - attention - sense - brother - special | 68 | 1439_kinda_attention_sense_brother |
| 1440 | games - playing - live - play - lots | 68 | 1440_games_playing_live_play |
| 1441 | excited - legit - fees - online - successful | 68 | 1441_excited_legit_fees_online |
| 1442 | similar - dude - better - true - sense | 67 | 1442_similar_dude_better_true |
| 1443 | updates - complete - minutes - specific - longer | 67 | 1443_updates_complete_minutes_specific |
| 1444 | fine - dev - specific - quick - example | 67 | 1444_fine_dev_specific_quick |
| 1445 | email - token - free - rate - main | 67 | 1445_email_token_free_rate |
| 1446 | previous - article - end - crazy - past | 67 | 1446_previous_article_end_crazy |
| 1447 | country - sir - help - nice - job | 67 | 1447_country_sir_help_nice |
| 1448 | date - contract - launch - website - project | 67 | 1448_date_contract_launch_website |
| 1449 | shit - play - game - dude - likely | 67 | 1449_shit_play_game_dude |
| 1450 | support - strong - major - level - likely | 67 | 1450_support_strong_major_level |
| 1451 | volume - increase - month - high - share | 67 | 1451_volume_increase_month_high |
| 1452 | chat - video - free - life - love | 67 | 1452_chat_video_free_life |
| 1453 | important - thing - dude - things - sense | 67 | 1453_important_thing_dude_things |
| 1454 | feature - turn - yesterday - end - ones | 67 | 1454_feature_turn_yesterday_end |
| 1455 | thinking - forward - potential - wallet - country | 67 | 1455_thinking_forward_potential_wallet |
| 1456 | network - smart - protocol - certain - block | 67 | 1456_network_smart_protocol_certain |
| 1457 | hours - new - days - release - development | 67 | 1457_hours_new_days_release |
| 1458 | rewards - nft - world - special - chance | 67 | 1458_rewards_nft_world_special |
| 1459 | opinion - app - visit - public - future | 67 | 1459_opinion_app_visit_public |
| 1460 | shit - type - positive - things - life | 67 | 1460_shit_type_positive_things |
| 1461 | thing - right - - - | 66 | 1461_thing_right__ |
| 1462 | situation - yesterday - bit - little - half | 66 | 1462_situation_yesterday_bit_little |
| 1463 | upcoming - block - blockchain - live - security | 66 | 1463_upcoming_block_blockchain_live |
| 1464 | product - release - stuff - bit - space | 66 | 1464_product_release_stuff_bit |
| 1465 | max - true - okay - possible - tho | 66 | 1465_max_true_okay_possible |
| 1466 | sure - upcoming - future - near - certain | 66 | 1466_sure_upcoming_future_near |
| 1467 | friend - family - security - worth - free | 66 | 1467_friend_family_security_worth |
| 1468 | web - technology - article - site - projects | 66 | 1468_web_technology_article_site |
| 1469 | protocol - listing - spot - specific - utc | 66 | 1469_protocol_listing_spot_specific |
| 1470 | correct - answer - yesterday - minutes - note | 66 | 1470_correct_answer_yesterday_minutes |
| 1471 | chance - digital - problems - users - complete | 66 | 1471_chance_digital_problems_users |
| 1472 | hold - financial - group - position - trade | 66 | 1472_hold_financial_group_position |
| 1473 | fine - tomorrow - day - nice - awesome | 65 | 1473_fine_tomorrow_day_nice |
| 1474 | damn - awesome - cool - research - features | 65 | 1474_damn_awesome_cool_research |
| 1475 | hand - sorry - previous - recent - yesterday | 65 | 1475_hand_sorry_previous_recent |
| 1476 | glad - brother - opinion - normal - awesome | 65 | 1476_glad_brother_opinion_normal |
| 1477 | country - likely - job - past - shit | 65 | 1477_country_likely_job_past |
| 1478 | date - active - available - - | 65 | 1478_date_active_available_ |
| 1479 | welcome - official - break - community - year | 65 | 1479_welcome_official_break_community |
| 1480 | solution - blockchain - technology - potential - problem | 65 | 1480_solution_blockchain_technology_potential |
| 1481 | special - service - year - play - payment | 65 | 1481_special_service_year_play |
| 1482 | country - technology - word - hand - near | 65 | 1482_country_technology_word_hand |
| 1483 | product - assets - digital - cryptocurrency - potential | 65 | 1483_product_assets_digital_cryptocurrency |
| 1484 | offer - limited - time - link - company | 65 | 1484_offer_limited_time_link |
| 1485 | attention - details - legit - right - platform | 65 | 1485_attention_details_legit_right |
| 1486 | ai - year - past - massive - share | 65 | 1486_ai_year_past_massive |
| 1487 | action - post - channel - user - pm | 65 | 1487_action_post_channel_user |
| 1488 | luck - family - ethereum - friends - coin | 65 | 1488_luck_family_ethereum_friends |
| 1489 | announcement - links - process - social - funds | 65 | 1489_announcement_links_process_social |
| 1490 | country - control - dude - world - certain | 65 | 1490_country_control_dude_world |
| 1491 | feature - announcement - features - dear - related | 65 | 1491_feature_announcement_features_dear |
| 1492 | turn - issue - cause - feature - issues | 65 | 1492_turn_issue_cause_feature |
| 1493 | list - personal - single - answer - run | 65 | 1493_list_personal_single_answer |
| 1494 | services - success - platform - right - investment | 65 | 1494_services_success_platform_right |
| 1495 | recent - glad - stuff - shit - huge | 65 | 1495_recent_glad_stuff_shit |
| 1496 | stuff - couple - control - fact - interesting | 64 | 1496_stuff_couple_control_fact |
| 1497 | media - social - active - big - money | 64 | 1497_media_social_active_big |
| 1498 | product - marketing - huge - single - sense | 64 | 1498_product_marketing_huge_single |
| 1499 | chart - use - trading - specific - deal | 64 | 1499_chart_use_trading_specific |
| 1500 | media - social - active - big - money | 64 | 1500_media_social_active_big |
| 1501 | new - dude - couple - guys - friends | 64 | 1501_new_dude_couple_guys |
| 1502 | past - future - tho - focus - wrong | 64 | 1502_past_future_tho_focus |
| 1503 | market - - - - | 64 | 1503_market___ |
| 1504 | nice - glad - night - fine - guys | 64 | 1504_nice_glad_night_fine |
| 1505 | live - guys - team - best - | 64 | 1505_live_guys_team_best |
| 1506 | country - job - interesting - shit - question | 64 | 1506_country_job_interesting_shit |
| 1507 | ones - hand - code - situation - glad | 64 | 1507_ones_hand_code_situation |
| 1508 | country - head - social - difference - clear | 64 | 1508_country_head_social_difference |
| 1509 | problem - love - glad - shit - guy | 64 | 1509_problem_love_glad_shit |
| 1510 | solution - list - telegram - best - huge | 64 | 1510_solution_list_telegram_best |
| 1511 | old - fact - cool - glad - interesting | 64 | 1511_old_fact_cool_glad |
| 1512 | matter - guide - version - reason - solution | 64 | 1512_matter_guide_version_reason |
| 1513 | excited - run - massive - glad - love | 64 | 1513_excited_run_massive_glad |
| 1514 | chain - wonder - moment - list - couple | 64 | 1514_chain_wonder_moment_list |
| 1515 | links - lots - high - active - page | 64 | 1515_links_lots_high_active |
| 1516 | trust - wallet - option - issues - guide | 64 | 1516_trust_wallet_option_issues |
| 1517 | previous - launch - supply - contract - transaction | 64 | 1517_previous_launch_supply_contract |
| 1518 | drop - date - pm - today - thanks | 63 | 1518_drop_date_pm_today |
| 1519 | perfect - easy - access - platform - thanks | 63 | 1519_perfect_easy_access_platform |
| 1520 | clear - tbh - interesting - cool - ecosystem | 63 | 1520_clear_tbh_interesting_cool |
| 1521 | feature - dev - main - step - stuff | 63 | 1521_feature_dev_main_step |
| 1522 | weeks - clear - certain - deal - case | 63 | 1522_weeks_clear_certain_deal |
| 1523 | damn - launch - community - - | 63 | 1523_damn_launch_community_ |
| 1524 | bot - cryptocurrency - market - profit - daily | 63 | 1524_bot_cryptocurrency_market_profit |
| 1525 | rate - quick - tomorrow - service - amazing | 63 | 1525_rate_quick_tomorrow_service |
| 1526 | listing - person - financial - potential - business | 63 | 1526_listing_person_financial_potential |
| 1527 | turn - pretty - fine - words - years | 63 | 1527_turn_pretty_fine_words |
| 1528 | chart - network - short - protocol - news | 63 | 1528_chart_network_short_protocol |
| 1529 | past - years - tbh - couple - likely | 63 | 1529_past_years_tbh_couple |
| 1530 | pepe - coin - coins - damn - run | 63 | 1530_pepe_coin_coins_damn |
| 1531 | dude - short - max - sir - normal | 63 | 1531_dude_short_max_sir |
| 1532 | risk - level - official - daily - | 63 | 1532_risk_level_official_daily |
| 1533 | high - big - time - - | 63 | 1533_high_big_time_ |
| 1534 | investing - mate - sure - market - space | 63 | 1534_investing_mate_sure_market |
| 1535 | dude - sense - news - fake - fair | 63 | 1535_dude_sense_news_fake |
| 1536 | different - plans - fees - online - way | 63 | 1536_different_plans_fees_online |
| 1537 | attention - questions - official - website - account | 63 | 1537_attention_questions_official_website |
| 1538 | stop - type - short - long - term | 63 | 1538_stop_type_short_long |
| 1539 | head - list - bad - point - weeks | 63 | 1539_head_list_bad_point |
| 1540 | games - play - game - order - interesting | 63 | 1540_games_play_game_order |
| 1541 | near - likely - lots - long - higher | 63 | 1541_near_likely_lots_long |
| 1542 | half - year - focus - fine - term | 63 | 1542_half_year_focus_fine |
| 1543 | position - place - lots - mind - guys | 63 | 1543_position_place_lots_mind |
| 1544 | shit - couple - close - alot - bad | 62 | 1544_shit_couple_close_alot |
| 1545 | welcome - free - payment - single - hour | 62 | 1545_welcome_free_payment_single |
| 1546 | important - buying - telegram - buy - example | 62 | 1546_important_buying_telegram_buy |
| 1547 | likely - interesting - end - things - late | 62 | 1547_likely_interesting_end_things |
| 1548 | control - point - position - second - shit | 62 | 1548_control_point_position_second |
| 1549 | feature - biggest - sure - project - matter | 62 | 1549_feature_biggest_sure_project |
| 1550 | face - awesome - dude - super - bro | 62 | 1550_face_awesome_dude_super |
| 1551 | code - dev - hard - ethereum - idea | 62 | 1551_code_dev_hard_ethereum |
| 1552 | event - contact - launch - email - latest | 62 | 1552_event_contact_launch_email |
| 1553 | clear - fact - different - face - lol | 62 | 1553_clear_fact_different_face |
| 1554 | term - fun - space - long - project | 62 | 1554_term_fun_space_long |
| 1555 | key - group - rules - moment - web | 62 | 1555_key_group_rules_moment |
| 1556 | hours - new - days - - | 62 | 1556_hours_new_days_ |
| 1557 | correct - game - problems - sense - answer | 62 | 1557_correct_game_problems_sense |
| 1558 | binance - usdt - sign - period - product | 62 | 1558_binance_usdt_sign_period |
| 1559 | brother - tbh - glad - yesterday - thanks | 62 | 1559_brother_tbh_glad_yesterday |
| 1560 | financial - today - platform - control - investment | 62 | 1560_financial_today_platform_control |
| 1561 | quick - open - limited - - | 62 | 1561_quick_open_limited_ |
| 1562 | feature - related - features - swap - dear | 62 | 1562_feature_related_features_swap |
| 1563 | problems - opinion - fair - public - media | 61 | 1563_problems_opinion_fair_public |
| 1564 | amazing - potential - market - important - security | 61 | 1564_amazing_potential_market_important |
| 1565 | sounds - phone - features - specific - different | 61 | 1565_sounds_phone_features_specific |
| 1566 | chat - interested - talk - meet - real | 61 | 1566_chat_interested_talk_meet |
| 1567 | step - world - term - amazing - glad | 61 | 1567_step_world_term_amazing |
| 1568 | note - site - plans - free - half | 61 | 1568_note_site_plans_free |
| 1569 | community - great - day - good - job | 61 | 1569_community_great_day_good |
| 1570 | ecosystem - article - type - works - product | 61 | 1570_ecosystem_article_type_works |
| 1571 | benefits - specific - certain - research - different | 61 | 1571_benefits_specific_certain_research |
| 1572 | late - private - channel - friends - dear | 61 | 1572_late_private_channel_friends |
| 1573 | point - comment - fine - answer - fun | 61 | 1573_point_comment_fine_answer |
| 1574 | opinion - idea - problems - strong - important | 61 | 1574_opinion_idea_problems_strong |
| 1575 | message - active - page - big - available | 61 | 1575_message_active_page_big |
| 1576 | admin - block - report - private - reply | 61 | 1576_admin_block_report_private |
| 1577 | guy - special - reply - real - talk | 61 | 1577_guy_special_reply_real |
| 1578 | form - night - game - dude - live | 61 | 1578_form_night_game_dude |
| 1579 | pool - utc - guide - end - period | 61 | 1579_pool_utc_guide_end |
| 1580 | pm - message - mind - answer - question | 61 | 1580_pm_message_mind_answer |
| 1581 | market - trading - trade - current - yesterday | 61 | 1581_market_trading_trade_current |
| 1582 | guy - dude - idea - awesome - correct | 61 | 1582_guy_dude_idea_awesome |
| 1583 | withdrawal - account - - - | 61 | 1583_withdrawal_account__ |
| 1584 | hours - new - days - comment - things | 61 | 1584_hours_new_days_comment |
| 1585 | support - sure - clear - reason - fact | 61 | 1585_support_sure_clear_reason |
| 1586 | difference - family - members - holding - guys | 61 | 1586_difference_family_members_holding |
| 1587 | word - answer - special - reply - fact | 61 | 1587_word_answer_special_reply |
| 1588 | damn - mate - ways - better - guy | 61 | 1588_damn_mate_ways_better |
| 1589 | - - - - | 61 | 1589____ |
| 1590 | digital - control - clear - solution - financial | 60 | 1590_digital_control_clear_solution |
| 1591 | web - solution - ecosystem - needs - users | 60 | 1591_web_solution_ecosystem_needs |
| 1592 | token - current - mate - step - launch | 60 | 1592_token_current_mate_step |
| 1593 | kind - team - people - stuff - lot | 60 | 1593_kind_team_people_stuff |
| 1594 | time - cool - place - mate - buddy | 60 | 1594_time_cool_place_mate |
| 1595 | small - yesterday - low - night - research | 60 | 1595_small_yesterday_low_night |
| 1596 | ta - product - lots - benefits - research | 60 | 1596_ta_product_lots_benefits |
| 1597 | phone - version - option - latest - global | 60 | 1597_phone_version_option_latest |
| 1598 | friend - true - legit - huge - thanks | 60 | 1598_friend_true_legit_huge |
| 1599 | real - work - example - recent - interesting | 60 | 1599_real_work_example_recent |
| 1600 | right - invest - time - late - early | 60 | 1600_right_invest_time_late |
| 1601 | level - bot - welcome - claim - hours | 60 | 1601_level_bot_welcome_claim |
| 1602 | country - job - reason - couple - brother | 60 | 1602_country_job_reason_couple |
| 1603 | event - project - team - offer - guys | 60 | 1603_event_project_team_offer |
| 1604 | order - short - profit - sell - bit | 60 | 1604_order_short_profit_sell |
| 1605 | fee - spot - head - article - fees | 59 | 1605_fee_spot_head_article |
| 1606 | fair - hour - normal - tho - buddy | 59 | 1606_fair_hour_normal_tho |
| 1607 | ways - interesting - stuff - shit - option | 59 | 1607_ways_interesting_stuff_shit |
| 1608 | important - success - profit - good - money | 59 | 1608_important_success_profit_good |
| 1609 | code - problems - issues - past - data | 59 | 1609_code_problems_issues_past |
| 1610 | things - possible - kind - visit - course | 59 | 1610_things_possible_kind_visit |
| 1611 | glad - form - reason - things - better | 59 | 1611_glad_form_reason_things |
| 1612 | issues - option - open - problems - opinion | 59 | 1612_issues_option_open_problems |
| 1613 | major - interesting - couple - longer - exciting | 59 | 1613_major_interesting_couple_longer |
| 1614 | legit - plan - little - profits - code | 59 | 1614_legit_plan_little_profits |
| 1615 | sense - lol - bro - buddy - alot | 59 | 1615_sense_lol_bro_buddy |
| 1616 | deposit - daily - term - minimum - level | 59 | 1616_deposit_daily_term_minimum |
| 1617 | moment - kind - huge - problems - withdrawal | 59 | 1617_moment_kind_huge_problems |
| 1618 | hand - security - head - project - cool | 59 | 1618_hand_security_head_project |
| 1619 | year - end - half - opinion - previous | 59 | 1619_year_end_half_opinion |
| 1620 | couple - issues - stuff - lots - certain | 59 | 1620_couple_issues_stuff_lots |
| 1621 | twitter - sounds - normal - fact - late | 59 | 1621_twitter_sounds_normal_fact |
| 1622 | real - start - couple - amazing - success | 59 | 1622_real_start_couple_amazing |
| 1623 | year - potential - page - work - | 59 | 1623_year_potential_page_work |
| 1624 | chat - term - real - long - fun | 59 | 1624_chat_term_real_long |
| 1625 | market - current - correct - needs - value | 59 | 1625_market_current_correct_needs |
| 1626 | couple - times - yesterday - sure - past | 59 | 1626_couple_times_yesterday_sure |
| 1627 | step - guide - mentioned - video - glad | 59 | 1627_step_guide_mentioned_video |
| 1628 | risk - level - official - daily - upcoming | 59 | 1628_risk_level_official_daily |
| 1629 | unique - lot - - - | 58 | 1629_unique_lot__ |
| 1630 | bad - difference - major - good - hour | 58 | 1630_bad_difference_major_good |
| 1631 | guy - friend - product - dm - fun | 58 | 1631_guy_friend_product_dm |
| 1632 | happy - life - today - good - investment | 58 | 1632_happy_life_today_good |
| 1633 | short - supply - protocol - total - | 58 | 1633_short_supply_protocol_total |
| 1634 | open - person - correct - guy - option | 58 | 1634_open_person_correct_guy |
| 1635 | shit - likely - ok - point - place | 58 | 1635_shit_likely_ok_point |
| 1636 | deposit - main - minimum - level - note | 58 | 1636_deposit_main_minimum_level |
| 1637 | face - build - phone - option - security | 58 | 1637_face_build_phone_option |
| 1638 | rest - higher - amazing - info - high | 58 | 1638_rest_higher_amazing_info |
| 1639 | network - building - protocol - things - solution | 58 | 1639_network_building_protocol_things |
| 1640 | possible - register - withdraw - investors - great | 58 | 1640_possible_register_withdraw_investors |
| 1641 | research - longer - bot - cause - buddy | 58 | 1641_research_longer_bot_cause |
| 1642 | second - volume - transactions - binance - issues | 58 | 1642_second_volume_transactions_binance |
| 1643 | example - set - list - claim - special | 58 | 1643_example_set_list_claim |
| 1644 | questions - ama - excited - interested - research | 58 | 1644_questions_ama_excited_interested |
| 1645 | accounts - visit - email - special - live | 58 | 1645_accounts_visit_email_special |
| 1646 | cryptocurrency - plans - exchange - security - services | 58 | 1646_cryptocurrency_plans_exchange_security |
| 1647 | available - info - time - - | 58 | 1647_available_info_time_ |
| 1648 | - - - - | 58 | 1648____ |
| 1649 | dude - previous - ama - buddy - weeks | 58 | 1649_dude_previous_ama_buddy |
| 1650 | data - key - couple - article - period | 58 | 1650_data_key_couple_article |
| 1651 | possible - - - - | 58 | 1651_possible___ |
| 1652 | doubt - platform - - - | 57 | 1652_doubt_platform__ |
| 1653 | word - attention - key - bit - thanks | 57 | 1653_word_attention_key_bit |
| 1654 | couple - lol - difference - example - ways | 57 | 1654_couple_lol_difference_example |
| 1655 | utc - period - app - listing - official | 57 | 1655_utc_period_app_listing |
| 1656 | launch - defi - ama - huge - media | 57 | 1656_launch_defi_ama_huge |
| 1657 | action - link - contact - - | 57 | 1657_action_link_contact_ |
| 1658 | buddy - reply - guys - man - cool | 57 | 1658_buddy_reply_guys_man |
| 1659 | guys - crazy - half - head - amazing | 57 | 1659_guys_crazy_half_head |
| 1660 | early - near - example - small - crazy | 57 | 1660_early_near_example_small |
| 1661 | stuff - welcome - open - list - specific | 57 | 1661_stuff_welcome_open_list |
| 1662 | buddy - fine - glad - ok - friends | 57 | 1662_buddy_fine_glad_ok |
| 1663 | matter - months - bitcoin - protocol - article | 57 | 1663_matter_months_bitcoin_protocol |
| 1664 | biggest - hand - problem - perfect - shit | 57 | 1664_biggest_hand_problem_perfect |
| 1665 | features - special - yesterday - plan - ways | 57 | 1665_features_special_yesterday_plan |
| 1666 | price - upcoming - claim - hour - listing | 57 | 1666_price_upcoming_claim_hour |
| 1667 | usdt - reward - rewards - worth - spot | 57 | 1667_usdt_reward_rewards_worth |
| 1668 | normal - version - update - build - near | 57 | 1668_normal_version_update_build |
| 1669 | option - swap - holders - tokens - token | 57 | 1669_option_swap_holders_tokens |
| 1670 | past - period - fact - example - making | 57 | 1670_past_period_fact_example |
| 1671 | app - pay - interested - number - drop | 57 | 1671_app_pay_interested_number |
| 1672 | important - plan - tokens - token - sir | 57 | 1672_important_plan_tokens_token |
| 1673 | talk - man - hard - real - good | 57 | 1673_talk_man_hard_real |
| 1674 | past - strong - level - key - couple | 57 | 1674_past_strong_level_key |
| 1675 | night - friends - head - hand - similar | 57 | 1675_night_friends_head_hand |
| 1676 | short - yesterday - guys - bad - single | 57 | 1676_short_yesterday_guys_bad |
| 1677 | web - chain - network - glad - recent | 57 | 1677_web_chain_network_glad |
| 1678 | ecosystem - strong - features - technology - number | 57 | 1678_ecosystem_strong_features_technology |
| 1679 | data - release - report - previous - drop | 57 | 1679_data_release_report_previous |
| 1680 | business - profit - week - investors - days | 57 | 1680_business_profit_week_investors |
| 1681 | awesome - crazy - stuff - shit - brother | 57 | 1681_awesome_crazy_stuff_shit |
| 1682 | country - problems - lots - nice - yesterday | 57 | 1682_country_problems_lots_nice |
| 1683 | certain - pool - release - option - increase | 57 | 1683_certain_pool_release_option |
| 1684 | reason - message - link - access - network | 57 | 1684_reason_message_link_access |
| 1685 | close - past - stuff - massive - minutes | 57 | 1685_close_past_stuff_massive |
| 1686 | latest - rewards - win - event - answer | 57 | 1686_latest_rewards_win_event |
| 1687 | case - use - hand - cool - type | 57 | 1687_case_use_hand_cool |
| 1688 | key - things - glad - positive - times | 57 | 1688_key_things_glad_positive |
| 1689 | problems - problem - moment - issues - situation | 57 | 1689_problems_problem_moment_issues |
| 1690 | normal - tbh - buddy - shit - cool | 56 | 1690_normal_tbh_buddy_shit |
| 1691 | page - links - perfect - years - better | 56 | 1691_page_links_perfect_years |
| 1692 | enjoy - life - moment - mate - dude | 56 | 1692_enjoy_life_moment_mate |
| 1693 | dude - guys - difference - break - buddy | 56 | 1693_dude_guys_difference_break |
| 1694 | close - problems - solution - research - multiple | 56 | 1694_close_problems_solution_research |
| 1695 | holders - technology - higher - alot - token | 56 | 1695_holders_technology_higher_alot |
| 1696 | action - message - reason - digital - official | 56 | 1696_action_message_reason_digital |
| 1697 | hours - new - days - space - unique | 56 | 1697_hours_new_days_space |
| 1698 | personal - business - development - issue - digital | 56 | 1698_personal_business_development_issue |
| 1699 | man - world - complete - moment - guy | 56 | 1699_man_world_complete_moment |
| 1700 | strong - coins - btc - days - | 56 | 1700_strong_coins_btc_days |
| 1701 | deposit - site - daily - minimum - withdrawal | 56 | 1701_deposit_site_daily_minimum |
| 1702 | night - tomorrow - friends - good - morning | 56 | 1702_night_tomorrow_friends_good |
| 1703 | ta - product - close - look - stuff | 56 | 1703_ta_product_close_look |
| 1704 | guy - job - watch - interesting - hand | 56 | 1704_guy_job_watch_interesting |
| 1705 | love - half - brother - crazy - shit | 56 | 1705_love_half_brother_crazy |
| 1706 | utc - trading - open - yesterday - tomorrow | 56 | 1706_utc_trading_open_yesterday |
| 1707 | likely - problem - issues - main - dude | 56 | 1707_likely_problem_issues_main |
| 1708 | dude - shit - guy - space - lol | 56 | 1708_dude_shit_guy_space |
| 1709 | protocol - cause - event - hand - short | 56 | 1709_protocol_cause_event_hand |
| 1710 | single - special - man - dude - buddy | 56 | 1710_single_special_man_dude |
| 1711 | normal - past - special - reply - fix | 56 | 1711_normal_past_special_reply |
| 1712 | benefits - friends - members - number - reward | 56 | 1712_benefits_friends_members_number |
| 1713 | night - good - rest - dear - nice | 56 | 1713_night_good_rest_dear |
| 1714 | dude - time - clear - past - stuff | 56 | 1714_dude_time_clear_past |
| 1715 | massive - case - daily - support - sense | 56 | 1715_massive_case_daily_support |
| 1716 | work - job - haha - tho - wonder | 56 | 1716_work_job_haha_tho |
| 1717 | coin - hours - new - days - shit | 56 | 1717_coin_hours_new_days |
| 1718 | country - talk - okay - bad - imagine | 56 | 1718_country_talk_okay_bad |
| 1719 | coin - hours - new - days - | 56 | 1719_coin_hours_new_days |
| 1720 | course - hope - sorry - correct - stuff | 56 | 1720_course_hope_sorry_correct |
| 1721 | report - week - level - bitcoin - price | 56 | 1721_report_week_level_bitcoin |
| 1722 | project - team - interesting - strong - great | 56 | 1722_project_team_interesting_strong |
| 1723 | chart - huge - bitcoin - - | 56 | 1723_chart_huge_bitcoin_ |
| 1724 | type - user - total - network - supply | 56 | 1724_type_user_total_network |
| 1725 | special - thinking - answer - dude - wrong | 55 | 1725_special_thinking_answer_dude |
| 1726 | kind - issue - contact - admin - funds | 55 | 1726_kind_issue_contact_admin |
| 1727 | defi - network - ecosystem - services - project | 55 | 1727_defi_network_ecosystem_services |
| 1728 | correct - man - alright - person - sorry | 55 | 1728_correct_man_alright_person |
| 1729 | number - data - low - available - | 55 | 1729_number_data_low_available |
| 1730 | assets - exchange - way - research - trade | 55 | 1730_assets_exchange_way_research |
| 1731 | attention - public - transactions - claim - tokens | 55 | 1731_attention_public_transactions_claim |
| 1732 | left - normal - word - bro - spot | 55 | 1732_left_normal_word_bro |
| 1733 | search - option - example - tho - alot | 55 | 1733_search_option_example_tho |
| 1734 | job - glad - biggest - things - buddy | 55 | 1734_job_glad_biggest_things |
| 1735 | break - idea - shit - half - mind | 55 | 1735_break_idea_shit_half |
| 1736 | smart - discord - transactions - chain - claim | 55 | 1736_smart_discord_transactions_chain |
| 1737 | excited - perfect - information - capital - investing | 55 | 1737_excited_perfect_information_capital |
| 1738 | report - case - financial - questions - ai | 55 | 1738_report_case_financial_questions |
| 1739 | active - page - account - - | 55 | 1739_active_page_account_ |
| 1740 | thanks - channel - link - free - money | 55 | 1740_thanks_channel_link_free |
| 1741 | late - solution - certain - order - important | 55 | 1741_late_solution_certain_order |
| 1742 | benefits - single - idea - sense - awesome | 55 | 1742_benefits_single_idea_sense |
| 1743 | ai - help - today - specific - information | 55 | 1743_ai_help_today_specific |
| 1744 | nice - friends - buddy - cool - country | 55 | 1744_nice_friends_buddy_cool |
| 1745 | bit - little - sir - huge - trade | 55 | 1745_bit_little_sir_huge |
| 1746 | sir - luck - interested - person - words | 55 | 1746_sir_luck_interested_person |
| 1747 | fix - comment - global - works - chance | 55 | 1747_fix_comment_global_works |
| 1748 | biggest - access - limited - company - capital | 55 | 1748_biggest_access_limited_company |
| 1749 | fair - awesome - dev - launch - smart | 55 | 1749_fair_awesome_dev_launch |
| 1750 | tho - dude - bit - guys - little | 55 | 1750_tho_dude_bit_guys |
| 1751 | correct - run - hard - major - thing | 55 | 1751_correct_run_hard_major |
| 1752 | tomorrow - person - real - successful - services | 55 | 1752_tomorrow_person_real_successful |
| 1753 | play - better - free - guys - deposit | 55 | 1753_play_better_free_guys |
| 1754 | country - cryptocurrency - crypto - services - solution | 55 | 1754_country_cryptocurrency_crypto_services |
| 1755 | volume - chart - liquidity - transactions - holders | 55 | 1755_volume_chart_liquidity_transactions |
| 1756 | job - moment - position - head - business | 55 | 1756_job_moment_position_head |
| 1757 | discord - pool - utc - exciting - protocol | 55 | 1757_discord_pool_utc_exciting |
| 1758 | problems - financial - thanks - personal - business | 55 | 1758_problems_financial_thanks_personal |
| 1759 | claim - welcome - withdrawal - level - free | 55 | 1759_claim_welcome_withdrawal_level |
| 1760 | fun - unique - chat - ready - dm | 55 | 1760_fun_unique_chat_ready |
| 1761 | - - - - | 54 | 1761____ |
| 1762 | chain - guide - form - second - total | 54 | 1762_chain_guide_form_second |
| 1763 | clear - dev - excited - term - feature | 54 | 1763_clear_dev_excited_term |
| 1764 | dear - personal - benefits - single - fun | 54 | 1764_dear_personal_benefits_single |
| 1765 | sense - dude - lot - cause - guy | 54 | 1765_sense_dude_lot_cause |
| 1766 | upcoming - private - type - chain - official | 54 | 1766_upcoming_private_type_chain |
| 1767 | chat - admin - couple - questions - question | 54 | 1767_chat_admin_couple_questions |
| 1768 | blockchain - smart - protocol - data - services | 54 | 1768_blockchain_smart_protocol_data |
| 1769 | bit - guess - luck - little - correct | 54 | 1769_bit_guess_luck_little |
| 1770 | - - - - | 54 | 1770____ |
| 1771 | special - financial - place - success - company | 54 | 1771_special_financial_place_success |
| 1772 | action - message - - - | 54 | 1772_action_message__ |
| 1773 | tho - kind - example - cool - country | 54 | 1773_tho_kind_example_cool |
| 1774 | moment - hand - half - longer - reply | 54 | 1774_moment_hand_half_longer |
| 1775 | sense - kind - fun - ways - buddy | 54 | 1775_sense_kind_fun_ways |
| 1776 | long - time - dude - couple - brother | 54 | 1776_long_time_dude_couple |
| 1777 | head - type - sign - financial - case | 54 | 1777_head_type_sign_financial |
| 1778 | claim - site - tokens - dear - open | 54 | 1778_claim_site_tokens_dear |
| 1779 | code - worth - token - reward - wallet | 54 | 1779_code_worth_token_reward |
| 1780 | dude - ta - protocol - shit - early | 54 | 1780_dude_ta_protocol_shit |
| 1781 | binance - services - exchange - security - search | 54 | 1781_binance_services_exchange_security |
| 1782 | block - transactions - control - short - user | 54 | 1782_block_transactions_control_short |
| 1783 | low - high - sell - hand - sense | 54 | 1783_low_high_sell_hand |
| 1784 | different - difference - time - correct - similar | 54 | 1784_different_difference_time_correct |
| 1785 | development - focus - matter - head - success | 54 | 1785_development_focus_matter_head |
| 1786 | volume - chart - price - market - buy | 54 | 1786_volume_chart_price_market |
| 1787 | power - cryptocurrency - strong - digital - transaction | 54 | 1787_power_cryptocurrency_strong_digital |
| 1788 | development - smart - web - strong - blockchain | 54 | 1788_development_smart_web_strong |
| 1789 | website - links - important - private - official | 54 | 1789_website_links_important_private |
| 1790 | dude - correct - moment - multiple - right | 54 | 1790_dude_correct_moment_multiple |
| 1791 | ta - guy - hand - point - close | 53 | 1791_ta_guy_hand_point |
| 1792 | daily - term - invest - long - profit | 53 | 1792_daily_term_invest_long |
| 1793 | win - chance - guys - word - minutes | 53 | 1793_win_chance_guys_word |
| 1794 | event - excited - head - space - unique | 53 | 1794_event_excited_head_space |
| 1795 | early - job - user - ready - project | 53 | 1795_early_job_user_ready |
| 1796 | job - strong - dev - community - great | 53 | 1796_job_strong_dev_community |
| 1797 | hour - times - service - days - telegram | 53 | 1797_hour_times_service_days |
| 1798 | buying - sir - similar - code - fee | 53 | 1798_buying_sir_similar_code |
| 1799 | country - normal - person - true - option | 53 | 1799_country_normal_person_true |
| 1800 | risk - simple - matter - community - single | 53 | 1800_risk_simple_matter_community |
| 1801 | awesome - ones - option - damn - games | 53 | 1801_awesome_ones_option_damn |
| 1802 | crypto - dude - specific - job - glad | 53 | 1802_crypto_dude_specific_job |
| 1803 | certain - pool - hand - likely - specific | 53 | 1803_certain_pool_hand_likely |
| 1804 | run - term - clear - solution - long | 53 | 1804_run_term_clear_solution |
| 1805 | important - fake - channel - safe - community | 53 | 1805_important_fake_channel_safe |
| 1806 | hand - normal - past - lol - bad | 53 | 1806_hand_normal_past_lol |
| 1807 | video - chat - meet - face - option | 53 | 1807_video_chat_meet_face |
| 1808 | rules - admin - bit - group - reason | 53 | 1808_rules_admin_bit_group |
| 1809 | accounts - pay - team - best - | 53 | 1809_accounts_pay_team_best |
| 1810 | discord - issues - questions - mate - buddy | 53 | 1810_discord_issues_questions_mate |
| 1811 | research - open - multiple - mate - near | 53 | 1811_research_open_multiple_mate |
| 1812 | space - projects - reply - awesome - interesting | 53 | 1812_space_projects_reply_awesome |
| 1813 | social - word - links - guys - brother | 53 | 1813_social_word_links_guys |
| 1814 | minimum - possible - hand - spot - feature | 53 | 1814_minimum_possible_hand_spot |
| 1815 | links - link - web - site - update | 53 | 1815_links_link_web_site |
| 1816 | late - successful - fee - capital - easy | 53 | 1816_late_successful_fee_capital |
| 1817 | share - - - - | 53 | 1817_share___ |
| 1818 | price - couple - selling - reason - spot | 53 | 1818_price_couple_selling_reason |
| 1819 | public - private - near - crazy - fair | 53 | 1819_public_private_near_crazy |
| 1820 | dev - fact - sorry - thought - friend | 53 | 1820_dev_fact_sorry_thought |
| 1821 | close - sure - second - quick - chance | 53 | 1821_close_sure_second_quick |
| 1822 | positive - course - mate - situation - things | 53 | 1822_positive_course_mate_situation |
| 1823 | message - fast - available - today - | 53 | 1823_message_fast_available_today |
| 1824 | cryptocurrency - crypto - exchange - clear - global | 53 | 1824_cryptocurrency_crypto_exchange_clear |
| 1825 | late - correct - answer - public - friend | 53 | 1825_late_correct_answer_public |
| 1826 | spot - listing - tho - yesterday - upcoming | 53 | 1826_spot_listing_tho_yesterday |
| 1827 | features - better - best - opinion - old | 53 | 1827_features_better_best_opinion |
| 1828 | thank - successful - earning - chain - happy | 53 | 1828_thank_successful_earning_chain |
| 1829 | release - case - trade - worth - earlier | 53 | 1829_release_case_trade_worth |
| 1830 | stuff - earlier - current - public - article | 53 | 1830_stuff_earlier_current_public |
| 1831 | night - public - latest - issues - update | 53 | 1831_night_public_latest_issues |
| 1832 | job - online - term - long - old | 53 | 1832_job_online_term_long |
| 1833 | awesome - announcement - details - okay - sense | 52 | 1833_awesome_announcement_details_okay |
| 1834 | video - official - sorry - shows - alot | 52 | 1834_video_official_sorry_shows |
| 1835 | important - news - date - dear - information | 52 | 1835_important_news_date_dear |
| 1836 | period - utc - spot - listing - upcoming | 52 | 1836_period_utc_spot_listing |
| 1837 | couple - head - run - super - difference | 52 | 1837_couple_head_run_super |
| 1838 | team - note - thanks - earlier - similar | 52 | 1838_team_note_thanks_earlier |
| 1839 | issue - liquidity - block - contact - token | 52 | 1839_issue_liquidity_block_contact |
| 1840 | dude - game - chance - question - ones | 52 | 1840_dude_game_chance_question |
| 1841 | near - interesting - year - issues - bit | 52 | 1841_near_interesting_year_issues |
| 1842 | payment - great - crypto - dm - available | 52 | 1842_payment_great_crypto_dm |
| 1843 | short - position - kind - close - point | 52 | 1843_short_position_kind_close |
| 1844 | ecosystem - live - today - free - wallet | 52 | 1844_ecosystem_live_today_free |
| 1845 | control - cause - brother - type - likely | 52 | 1845_control_cause_brother_type |
| 1846 | crazy - idea - yesterday - block - way | 52 | 1846_crazy_idea_yesterday_block |
| 1847 | questions - minimum - years - email - information | 52 | 1847_questions_minimum_years_email |
| 1848 | simple - massive - pool - ecosystem - total | 52 | 1848_simple_massive_pool_ecosystem |
| 1849 | ones - half - months - couple - normal | 52 | 1849_ones_half_months_couple |
| 1850 | huge - successful - high - profit - investors | 52 | 1850_huge_successful_high_profit |
| 1851 | assets - blockchain - network - capital - trade | 52 | 1851_assets_blockchain_network_capital |
| 1852 | number - block - earlier - early - site | 52 | 1852_number_block_earlier_early |
| 1853 | event - project - team - yesterday - cool | 52 | 1853_event_project_team_yesterday |
| 1854 | increase - half - certain - clear - wonder | 52 | 1854_increase_half_certain_clear |
| 1855 | private - public - question - coins - point | 52 | 1855_private_public_question_coins |
| 1856 | media - social - limited - - | 52 | 1856_media_social_limited_ |
| 1857 | things - financial - future - token - life | 52 | 1857_things_financial_future_token |
| 1858 | site - video - day - new - daily | 52 | 1858_site_video_day_new |
| 1859 | higher - difference - similar - pool - protocol | 52 | 1859_higher_difference_similar_pool |
| 1860 | hand - wallet - half - worth - low | 52 | 1860_hand_wallet_half_worth |
| 1861 | shit - dude - buddy - guy - issues | 52 | 1861_shit_dude_buddy_guy |
| 1862 | worth - ethereum - hours - trade - withdrawal | 52 | 1862_worth_ethereum_hours_trade |
| 1863 | action - link - community - - | 52 | 1863_action_link_community_ |
| 1864 | high - future - tokens - matter - price | 52 | 1864_high_future_tokens_matter |
| 1865 | latest - version - live - upcoming - game | 52 | 1865_latest_version_live_upcoming |
| 1866 | account - admin - half - problem - issue | 52 | 1866_account_admin_half_problem |
| 1867 | article - guys - work - - | 52 | 1867_article_guys_work_ |
| 1868 | type - tokens - issues - protocol - block | 52 | 1868_type_tokens_issues_protocol |
| 1869 | chart - previous - higher - clear - action | 52 | 1869_chart_previous_higher_clear |
| 1870 | ama - yesterday - questions - chance - video | 52 | 1870_ama_yesterday_questions_chance |
| 1871 | blockchain - digital - comment - latest - technology | 52 | 1871_blockchain_digital_comment_latest |
| 1872 | web - dev - volume - token - daily | 52 | 1872_web_dev_volume_token |
| 1873 | year - amazing - info - fast - account | 52 | 1873_year_amazing_info_fast |
| 1874 | financial - successful - right - work - platform | 51 | 1874_financial_successful_right_work |
| 1875 | project - pool - event - trading - perfect | 51 | 1875_project_pool_event_trading |
| 1876 | biggest - chart - position - buy - hours | 51 | 1876_biggest_chart_position_buy |
| 1877 | rest - couple - yesterday - fair - positive | 51 | 1877_rest_couple_yesterday_fair |
| 1878 | pool - reward - web - version - friend | 51 | 1878_pool_reward_web_version |
| 1879 | major - bit - fees - rewards - action | 51 | 1879_major_bit_fees_rewards |
| 1880 | holders - happy - long - day - | 51 | 1880_holders_happy_long_day |
| 1881 | financial - global - true - longer - future | 51 | 1881_financial_global_true_longer |
| 1882 | shit - friend - difference - major - year | 51 | 1882_shit_friend_difference_major |
| 1883 | ama - global - follow - rewards - social | 51 | 1883_ama_global_follow_rewards |
| 1884 | face - awesome - things - happy - platform | 51 | 1884_face_awesome_things_happy |
| 1885 | new - head - public - awesome - word | 51 | 1885_new_head_public_awesome |
| 1886 | fact - positive - success - mind - times | 51 | 1886_fact_positive_success_mind |
| 1887 | simple - home - looks - mind - clear | 51 | 1887_simple_home_looks_mind |
| 1888 | brother - tomorrow - related - stuff - fine | 51 | 1888_brother_tomorrow_related_stuff |
| 1889 | ai - assets - utc - announcement - games | 51 | 1889_ai_assets_utc_announcement |
| 1890 | technology - real - world - digital - unique | 51 | 1890_technology_real_world_digital |
| 1891 | help - - - - | 51 | 1891_help___ |
| 1892 | market - period - couple - man - late | 51 | 1892_market_period_couple_man |
| 1893 | weeks - pepe - short - sense - profits | 51 | 1893_weeks_pepe_short_sense |
| 1894 | fair - current - transaction - launch - swap | 51 | 1894_fair_current_transaction_launch |
| 1895 | updates - bot - information - list - admin | 51 | 1895_updates_bot_information_list |
| 1896 | home - dude - head - article - shit | 51 | 1896_home_dude_head_article |
| 1897 | version - release - security - number - public | 51 | 1897_version_release_security_number |
| 1898 | couple - brother - thing - bad - worth | 51 | 1898_couple_brother_thing_bad |
| 1899 | wanted - guy - dude - lol - comment | 51 | 1899_wanted_guy_dude_lol |
| 1900 | ai - development - right - global - best | 51 | 1900_ai_development_right_global |
| 1901 | thing - best - right - man - smart | 51 | 1901_thing_best_right_man |
| 1902 | links - blockchain - information - contract - place | 51 | 1902_links_blockchain_information_contract |
| 1903 | block - message - user - public - private | 51 | 1903_block_message_user_public |
| 1904 | break - opportunity - special - yesterday - correct | 51 | 1904_break_opportunity_special_yesterday |
| 1905 | stop - position - close - order - short | 51 | 1905_stop_position_close_order |
| 1906 | dev - early - super - strong - biggest | 51 | 1906_dev_early_super_strong |
| 1907 | private - dear - fake - place - accounts | 51 | 1907_private_dear_fake_place |
| 1908 | friend - possible - hard - success - happy | 51 | 1908_friend_possible_hard_success |
| 1909 | fine - crypto - security - biggest - clear | 51 | 1909_fine_crypto_security_biggest |
| 1910 | form - block - transaction - payment - safe | 51 | 1910_form_block_transaction_payment |
| 1911 | development - potential - features - site - unique | 50 | 1911_development_potential_features_site |
| 1912 | difference - similar - dude - sense - lol | 50 | 1912_difference_similar_dude_sense |
| 1913 | usdt - nft - main - thought - level | 50 | 1913_usdt_nft_main_thought |
| 1914 | glad - sir - man - team - channel | 50 | 1914_glad_sir_man_team |
| 1915 | version - release - security - note - number | 50 | 1915_version_release_security_note |
| 1916 | week - special - excited - night - recent | 50 | 1916_week_special_excited_night |
| 1917 | cryptocurrency - second - market - today - normal | 50 | 1917_cryptocurrency_second_market_today |
| 1918 | similar - ai - bit - times - chat | 50 | 1918_similar_ai_bit_times |
| 1919 | ecosystem - network - live - product - token | 50 | 1919_ecosystem_network_live_product |
| 1920 | position - order - normal - trading - difference | 50 | 1920_position_order_normal_trading |
| 1921 | page - day - free - dm - total | 50 | 1921_page_day_free_dm |
| 1922 | higher - yesterday - true - certain - situation | 50 | 1922_higher_yesterday_true_certain |
| 1923 | half - cool - hour - important - details | 50 | 1923_half_cool_hour_important |
| 1924 | site - minutes - form - difference - kinda | 50 | 1924_site_minutes_form_difference |
| 1925 | transactions - control - dev - links - data | 50 | 1925_transactions_control_dev_links |
| 1926 | comment - video - digital - development - latest | 50 | 1926_comment_video_digital_development |
| 1927 | recent - positive - token - lots - ecosystem | 50 | 1927_recent_positive_token_lots |
| 1928 | control - ways - wrong - difference - interesting | 50 | 1928_control_ways_wrong_difference |
| 1929 | smart - technology - online - unique - business | 50 | 1929_smart_technology_online_unique |
| 1930 | weeks - page - dm - better - days | 50 | 1930_weeks_page_dm_better |
| 1931 | info - years - fast - - | 50 | 1931_info_years_fast_ |
| 1932 | announcement - wrong - pool - list - weeks | 50 | 1932_announcement_wrong_pool_list |
| 1933 | excited - payment - type - wallet - friend | 50 | 1933_excited_payment_type_wallet |
| 1934 | stop - likely - idea - lol - stuff | 50 | 1934_stop_likely_idea_lol |
| 1935 | sense - opinion - clear - interesting - correct | 50 | 1935_sense_opinion_clear_interesting |
| 1936 | plans - fee - years - trading - register | 50 | 1936_plans_fee_years_trading |
| 1937 | cool - mate - sure - shit - bro | 50 | 1937_cool_mate_sure_shit |
| 1938 | rest - work - account - minimum - hard | 50 | 1938_rest_work_account_minimum |
| 1939 | order - things - success - life - key | 50 | 1939_order_things_success_life |
| 1940 | telegram - ways - situation - personal - person | 50 | 1940_telegram_ways_situation_personal |
| 1941 | code - choose - unique - quick - friend | 50 | 1941_code_choose_unique_quick |
| 1942 | similar - higher - swap - page - dm | 50 | 1942_similar_higher_swap_page |
| 1943 | protocol - clear - night - place - exciting | 50 | 1943_protocol_clear_night_place |
| 1944 | selling - sell - buying - half - early | 50 | 1944_selling_sell_buying_half |
| 1945 | comment - example - solution - network - protocol | 50 | 1945_comment_example_solution_network |
| 1946 | year - excited - note - opinion - buddy | 50 | 1946_year_excited_note_opinion |
| 1947 | early - end - previous - interesting - moment | 50 | 1947_early_end_previous_interesting |
| 1948 | rules - links - group - kind - chat | 50 | 1948_rules_links_group_kind |
| 1949 | total - claim - value - unique - assets | 49 | 1949_total_claim_value_unique |
| 1950 | event - - - - | 49 | 1950_event___ |
| 1951 | welcome - exchange - form - volume - difference | 49 | 1951_welcome_exchange_form_volume |
| 1952 | rules - announcement - official - welcome - chat | 49 | 1952_rules_announcement_official_welcome |
| 1953 | media - social - limited - - | 49 | 1953_media_social_limited_ |
| 1954 | usdt - complete - friends - rewards - users | 49 | 1954_usdt_complete_friends_rewards |
| 1955 | things - similar - thing - fact - sure | 49 | 1955_things_similar_thing_fact |
| 1956 | course - action - control - hand - form | 49 | 1956_course_action_control_hand |
| 1957 | shit - ama - benefits - article - tomorrow | 49 | 1957_shit_ama_benefits_article |
| 1958 | help - mate - sir - sure - great | 49 | 1958_help_mate_sir_sure |
| 1959 | shows - live - feature - rate - selling | 49 | 1959_shows_live_feature_rate |
| 1960 | fees - eth - fee - crazy - option | 49 | 1960_fees_eth_fee_crazy |
| 1961 | clear - major - moment - btc - case | 49 | 1961_clear_major_moment_btc |
| 1962 | public - coins - cryptocurrency - years - bitcoin | 49 | 1962_public_coins_cryptocurrency_years |
| 1963 | coin - hours - new - - | 49 | 1963_coin_hours_new_ |
| 1964 | fun - opinion - stuff - moment - lot | 49 | 1964_fun_opinion_stuff_moment |
| 1965 | bit - day - break - rest - fine | 49 | 1965_bit_day_break_rest |
| 1966 | future - features - version - simple - updates | 49 | 1966_future_features_version_simple |
| 1967 | bad - issue - media - social - opinion | 49 | 1967_bad_issue_media_social |
| 1968 | feature - latest - idea - user - bad | 49 | 1968_feature_latest_idea_user |
| 1969 | reply - answer - question - ones - idea | 49 | 1969_reply_answer_question_ones |
| 1970 | reason - course - dude - biggest - night | 49 | 1970_reason_course_dude_biggest |
| 1971 | list - years - contract - bitcoin - deposit | 49 | 1971_list_years_contract_bitcoin |
| 1972 | trade - - - - | 49 | 1972_trade___ |
| 1973 | accounts - group - half - account - trading | 49 | 1973_accounts_group_half_account |
| 1974 | transaction - swap - fees - worth - example | 49 | 1974_transaction_swap_fees_worth |
| 1975 | period - ethereum - ones - term - coins | 49 | 1975_period_ethereum_ones_term |
| 1976 | supply - total - worth - tokens - liquidity | 49 | 1976_supply_total_worth_tokens |
| 1977 | interesting - volume - period - number - sense | 49 | 1977_interesting_volume_period_number |
| 1978 | massive - dude - opinion - wrong - biggest | 49 | 1978_massive_dude_opinion_wrong |
| 1979 | moment - person - single - brother - buddy | 49 | 1979_moment_person_single_brother |
| 1980 | minimum - game - app - high - withdraw | 49 | 1980_minimum_game_app_high |
| 1981 | morning - today - good - day - buddy | 49 | 1981_morning_today_good_day |
| 1982 | details - plan - hand - case - opinion | 49 | 1982_details_plan_hand_case |
| 1983 | good - extra - - - | 49 | 1983_good_extra__ |
| 1984 | successful - big - group - version - | 49 | 1984_successful_big_group_version |
| 1985 | amazing - success - opportunity - capital - life | 49 | 1985_amazing_success_opportunity_capital |
| 1986 | minutes - block - total - reward - rewards | 49 | 1986_minutes_block_total_reward |
| 1987 | available - active - message - - | 49 | 1987_available_active_message_ |
| 1988 | price - binance - token - date - increase | 49 | 1988_price_binance_token_date |
| 1989 | article - week - option - main - family | 49 | 1989_article_week_option_main |
| 1990 | dude - couple - ok - sense - normal | 49 | 1990_dude_couple_ok_sense |
| 1991 | change - cause - thing - question - shit | 49 | 1991_change_cause_thing_question |
| 1992 | money - interested - business - exchange - available | 48 | 1992_money_interested_business_exchange |
| 1993 | weeks - months - dude - sorry - break | 48 | 1993_weeks_months_dude_sorry |
| 1994 | news - care - hand - head - left | 48 | 1994_news_care_hand_head |
| 1995 | super - specific - option - feature - app | 48 | 1995_super_specific_option_feature |
| 1996 | longer - type - recent - block - lots | 48 | 1996_longer_type_recent_block |
| 1997 | half - similar - home - ready - today | 48 | 1997_half_similar_home_ready |
| 1998 | massive - smart - launch - transaction - liquidity | 48 | 1998_massive_smart_launch_transaction |
| 1999 | hand - week - information - works - admin | 48 | 1999_hand_week_information_works |
| 2000 | binance - dm - - - | 48 | 2000_binance_dm__ |
| 2001 | coin - hours - new - days - | 48 | 2001_coin_hours_new_days |
| 2002 | group - action - special - second - reply | 48 | 2002_group_action_special_second |
| 2003 | multiple - hour - info - services - channel | 48 | 2003_multiple_hour_info_services |
| 2004 | security - issues - worth - digital - trust | 48 | 2004_security_issues_worth_digital |
| 2005 | pepe - brother - fair - mind - space | 48 | 2005_pepe_brother_fair_mind |
| 2006 | reply - specific - case - answer - brother | 48 | 2006_reply_specific_case_answer |
| 2007 | head - issues - different - moment - likely | 48 | 2007_head_issues_different_moment |
| 2008 | fix - problems - available - accounts - phone | 48 | 2008_fix_problems_available_accounts |
| 2009 | report - phone - info - business - number | 48 | 2009_report_phone_info_business |
| 2010 | rest - work - account - minimum - hard | 48 | 2010_rest_work_account_minimum |
| 2011 | bot - trading - spot - rate - code | 48 | 2011_bot_trading_spot_rate |
| 2012 | page - extra - payment - dm - free | 48 | 2012_page_extra_payment_dm |
| 2013 | code - enter - app - email - register | 48 | 2013_code_enter_app_email |
| 2014 | hour - post - big - - | 48 | 2014_hour_post_big_ |
| 2015 | glad - man - amazing - financial - issues | 48 | 2015_glad_man_amazing_financial |
| 2016 | dude - nice - short - glad - shit | 48 | 2016_dude_nice_short_glad |
| 2017 | service - services - create - level - users | 48 | 2017_service_services_create_level |
| 2018 | little - quick - private - difference - example | 48 | 2018_little_quick_private_difference |
| 2019 | tho - point - example - interesting - simple | 48 | 2019_tho_point_example_interesting |
| 2020 | coins - coin - value - sense - fact | 48 | 2020_coins_coin_value_sense |
| 2021 | eth - btc - case - focus - shit | 48 | 2021_eth_btc_case_focus |
| 2022 | account - company - withdrawal - bitcoin - good | 48 | 2022_account_company_withdrawal_bitcoin |
| 2023 | lot - specific - lots - recent - number | 48 | 2023_lot_specific_lots_recent |
| 2024 | code - claim - binance - crypto - free | 48 | 2024_code_claim_binance_crypto |
| 2025 | multiple - benefits - launch - supply - marketing | 48 | 2025_multiple_benefits_launch_supply |
| 2026 | half - hour - morning - break - little | 48 | 2026_half_hour_morning_break |
| 2027 | projects - development - upcoming - technology - features | 48 | 2027_projects_development_upcoming_technology |
| 2028 | hand - personal - profit - trading - amazing | 48 | 2028_hand_personal_profit_trading |
| 2029 | upcoming - specific - network - ecosystem - excited | 48 | 2029_upcoming_specific_network_ecosystem |
| 2030 | smart - article - contract - interesting - code | 48 | 2030_smart_article_contract_interesting |
| 2031 | page - big - channel - available - specific | 48 | 2031_page_big_channel_available |
| 2032 | single - love - dude - brother - late | 48 | 2032_single_love_dude_brother |
| 2033 | dude - course - moment - cool - bro | 48 | 2033_dude_course_moment_cool |
| 2034 | sir - yesterday - moment - feature - dear | 48 | 2034_sir_yesterday_moment_feature |
| 2035 | long - taking - ethereum - wait - hours | 48 | 2035_long_taking_ethereum_wait |
| 2036 | usdt - eth - ethereum - chain - address | 48 | 2036_usdt_eth_ethereum_chain |
| 2037 | dev - wonder - specific - brother - news | 48 | 2037_dev_wonder_specific_brother |
| 2038 | rate - high - step - opinion - point | 48 | 2038_rate_high_step_opinion |
| 2039 | ask - service - phone - security - free | 48 | 2039_ask_service_phone_security |
| 2040 | couple - worth - little - control - amazing | 48 | 2040_couple_worth_little_control |
| 2041 | love - person - kind - minutes - reason | 48 | 2041_love_person_kind_minutes |
| 2042 | damn - nice - tho - close - weeks | 47 | 2042_damn_nice_tho_close |
| 2043 | data - key - ethereum - price - btc | 47 | 2043_data_key_ethereum_price |
| 2044 | personal - date - private - media - social | 47 | 2044_personal_date_private_media |
| 2045 | fast - - - - | 47 | 2045_fast___ |
| 2046 | total - latest - supply - value - days | 47 | 2046_total_latest_supply_value |
| 2047 | unique - interesting - nice - fact - buddy | 47 | 2047_unique_interesting_nice_fact |
| 2048 | eth - pls - minutes - yesterday - correct | 47 | 2048_eth_pls_minutes_yesterday |
| 2049 | case - huge - worth - possible - care | 47 | 2049_case_huge_worth_possible |
| 2050 | utc - data - times - days - | 47 | 2050_utc_data_times_days |
| 2051 | time - good - tho - early - morning | 47 | 2051_time_good_tho_early |
| 2052 | tokens - release - rest - period - guide | 47 | 2052_tokens_release_rest_period |
| 2053 | minutes - group - admin - community - | 47 | 2053_minutes_group_admin_community |
| 2054 | rest - term - info - high - long | 47 | 2054_rest_term_info_high |
| 2055 | night - morning - mate - good - rest | 47 | 2055_night_morning_mate_good |
| 2056 | game - minutes - number - low - small | 47 | 2056_game_minutes_number_low |
| 2057 | rate - available - services - secure - coins | 47 | 2057_rate_available_services_secure |
| 2058 | needs - action - issues - hope - space | 47 | 2058_needs_action_issues_hope |
| 2059 | version - latest - guys - project - team | 47 | 2059_version_latest_guys_project |
| 2060 | important - control - biggest - turn - buddy | 47 | 2060_important_control_biggest_turn |
| 2061 | minimum - rules - utc - event - pool | 47 | 2061_minimum_rules_utc_event |
| 2062 | past - clear - half - dev - community | 47 | 2062_past_clear_half_dev |
| 2063 | tho - announcement - moment - interested - business | 47 | 2063_tho_announcement_moment_interested |
| 2064 | links - questions - discord - site - higher | 47 | 2064_links_questions_discord_site |
| 2065 | bot - free - friends - hours - profit | 47 | 2065_bot_free_friends_hours |
| 2066 | imo - things - future - situation - article | 47 | 2066_imo_things_future_situation |
| 2067 | product - couple - yesterday - hand - biggest | 47 | 2067_product_couple_yesterday_hand |
| 2068 | release - product - game - games - members | 47 | 2068_release_product_game_games |
| 2069 | weeks - guy - worth - single - hour | 47 | 2069_weeks_guy_worth_single |
| 2070 | awesome - quick - secure - safe - happy | 47 | 2070_awesome_quick_secure_safe |
| 2071 | app - strong - questions - links - official | 47 | 2071_app_strong_questions_links |
| 2072 | buddy - chat - specific - friend - guy | 47 | 2072_buddy_chat_specific_friend |
| 2073 | games - normal - best - tho - gives | 47 | 2073_games_normal_best_tho |
| 2074 | situation - btc - search - position - current | 47 | 2074_situation_btc_search_position |
| 2075 | app - visit - version - global - strong | 47 | 2075_app_visit_version_global |
| 2076 | google - ai - success - service - bot | 47 | 2076_google_ai_success_service |
| 2077 | event - technology - special - major - total | 47 | 2077_event_technology_special_major |
| 2078 | attention - huge - members - project - job | 47 | 2078_attention_huge_members_project |
| 2079 | process - step - certain - article - buddy | 47 | 2079_process_step_certain_article |
| 2080 | specific - services - media - social - rate | 47 | 2080_specific_services_media_social |
| 2081 | hour - second - friends - chance - family | 47 | 2081_hour_second_friends_chance |
| 2082 | bit - little - sense - head - wrong | 47 | 2082_bit_little_sense_head |
| 2083 | option - cause - thanks - latest - problem | 46 | 2083_option_cause_thanks_latest |
| 2084 | technology - blockchain - plans - digital - financial | 46 | 2084_technology_blockchain_plans_digital |
| 2085 | crypto - ways - similar - earning - members | 46 | 2085_crypto_ways_similar_earning |
| 2086 | potential - financial - plan - account - | 46 | 2086_potential_financial_plan_account |
| 2087 | massive - ethereum - welcome - future - public | 46 | 2087_massive_ethereum_welcome_future |
| 2088 | connect - option - site - buying - feature | 46 | 2088_connect_option_site_buying |
| 2089 | job - tho - yesterday - guy - friend | 46 | 2089_job_tho_yesterday_guy |
| 2090 | guy - tbh - shit - imagine - difference | 46 | 2090_guy_tbh_shit_imagine |
| 2091 | legit - fake - fees - extra - free | 46 | 2091_legit_fake_fees_extra |
| 2092 | pepe - bot - welcome - free - minimum | 46 | 2092_pepe_bot_welcome_free |
| 2093 | sounds - case - order - action - recent | 46 | 2093_sounds_case_order_action |
| 2094 | minimum - free - app - daily - level | 46 | 2094_minimum_free_app_daily |
| 2095 | easy - term - minutes - hard - site | 46 | 2095_easy_term_minutes_hard |
| 2096 | nft - tokens - product - different - token | 46 | 2096_nft_tokens_product_different |
| 2097 | tho - couple - end - small - yesterday | 46 | 2097_tho_couple_end_small |
| 2098 | pm - - - - | 46 | 2098_pm___ |
| 2099 | feature - trade - features - fees - experience | 46 | 2099_feature_trade_features_fees |
| 2100 | article - guys - weeks - coins - announcement | 46 | 2100_article_guys_weeks_coins |
| 2101 | fun - details - meet - strong - high | 46 | 2101_fun_details_meet_strong |
| 2102 | month - happy - morning - night - new | 46 | 2102_month_happy_morning_night |
| 2103 | investment - company - point - claim - fees | 46 | 2103_investment_company_point_claim |
| 2104 | services - capital - message - website - investors | 46 | 2104_services_capital_message_website |
| 2105 | risk - level - official - daily - | 46 | 2105_risk_level_official_daily |
| 2106 | development - experience - web - years - interested | 46 | 2106_development_experience_web_years |
| 2107 | wrong - question - showing - issue - ones | 46 | 2107_wrong_question_showing_issue |
| 2108 | half - article - place - fact - news | 46 | 2108_half_article_place_fact |
| 2109 | option - better - fact - solution - crazy | 46 | 2109_option_better_fact_solution |
| 2110 | channel - block - support - welcome - news | 46 | 2110_channel_block_support_welcome |
| 2111 | pool - tho - huge - event - earlier | 46 | 2111_pool_tho_huge_event |
| 2112 | upcoming - announcement - awesome - technology - projects | 46 | 2112_upcoming_announcement_awesome_technology |
| 2113 | hand - rest - article - fact - post | 46 | 2113_hand_rest_article_fact |
| 2114 | rate - coins - coin - sell - interested | 46 | 2114_rate_coins_coin_sell |
| 2115 | ready - day - good - - | 46 | 2115_ready_day_good_ |
| 2116 | ama - upcoming - chain - updates - questions | 46 | 2116_ama_upcoming_chain_updates |
| 2117 | massive - ama - launch - listing - potential | 46 | 2117_massive_ama_launch_listing |
| 2118 | risk - level - official - daily - | 46 | 2118_risk_level_official_daily |
| 2119 | members - user - binance - team - exchange | 46 | 2119_members_user_binance_team |
| 2120 | report - number - transactions - year - crypto | 46 | 2120_report_number_transactions_year |
| 2121 | face - short - real - life - mate | 46 | 2121_face_short_real_life |
| 2122 | nice - project - break - good - time | 46 | 2122_nice_project_break_good |
| 2123 | hand - course - special - normal - night | 46 | 2123_hand_course_special_normal |
| 2124 | country - person - ok - difference - glad | 46 | 2124_country_person_ok_difference |
| 2125 | sorry - legit - man - guys - group | 46 | 2125_sorry_legit_man_guys |
| 2126 | situation - available - bad - thing - yesterday | 46 | 2126_situation_available_bad_thing |
| 2127 | data - app - links - glad - cool | 46 | 2127_data_app_links_glad |
| 2128 | bit - point - person - little - easy | 46 | 2128_bit_point_person_little |
| 2129 | code - term - long - project - link | 45 | 2129_code_term_long_project |
| 2130 | issue - brother - update - phone - solution | 45 | 2130_issue_brother_update_phone |
| 2131 | claim - withdrawal - new - - | 45 | 2131_claim_withdrawal_new_ |
| 2132 | minutes - trade - ones - profit - profits | 45 | 2132_minutes_trade_ones_profit |
| 2133 | research - value - cryptocurrency - rate - important | 45 | 2133_research_value_cryptocurrency_rate |
| 2134 | ecosystem - tokens - attention - exciting - value | 45 | 2134_ecosystem_tokens_attention_exciting |
| 2135 | block - moment - supply - recent - current | 45 | 2135_block_moment_supply_recent |
| 2136 | specific - answer - multiple - bit - things | 45 | 2136_specific_answer_multiple_bit |
| 2137 | job - interested - online - hour - week | 45 | 2137_job_interested_online_hour |
| 2138 | chat - message - bot - - | 45 | 2138_chat_message_bot_ |
| 2139 | type - word - option - web - google | 45 | 2139_type_word_option_web |
| 2140 | feature - chance - luck - weeks - extra | 45 | 2140_feature_chance_luck_weeks |
| 2141 | trading - successful - previous - option - place | 45 | 2141_trading_successful_previous_option |
| 2142 | tomorrow - wallet - thanks - question - hours | 45 | 2142_tomorrow_wallet_thanks_question |
| 2143 | upcoming - welcome - old - people - | 45 | 2143_upcoming_welcome_old_people |
| 2144 | code - friend - issue - app - key | 45 | 2144_code_friend_issue_app |
| 2145 | sense - benefits - guess - web - chain | 45 | 2145_sense_benefits_guess_web |
| 2146 | lots - links - high - active - big | 45 | 2146_lots_links_high_active |
| 2147 | near - brother - ways - word - bit | 45 | 2147_near_brother_ways_word |
| 2148 | idea - post - brother - mind - awesome | 45 | 2148_idea_post_brother_mind |
| 2149 | - - - - | 45 | 2149____ |
| 2150 | global - main - feature - version - latest | 45 | 2150_global_main_feature_version |
| 2151 | past - late - break - run - couple | 45 | 2151_past_late_break_run |
| 2152 | protocol - release - liquidity - benefits - increase | 45 | 2152_protocol_release_liquidity_benefits |
| 2153 | games - game - excited - attention - binance | 45 | 2153_games_game_excited_attention |
| 2154 | buddy - hand - previous - bro - cause | 45 | 2154_buddy_hand_previous_bro |
| 2155 | positive - life - fact - happy - way | 45 | 2155_positive_life_fact_happy |
| 2156 | games - game - cause - strong - man | 45 | 2156_games_game_cause_strong |
| 2157 | massive - bit - feature - yesterday - interesting | 45 | 2157_massive_bit_feature_yesterday |
| 2158 | volume - change - low - high - price | 45 | 2158_volume_change_low_high |
| 2159 | questions - glad - question - post - | 45 | 2159_questions_glad_question_post |
| 2160 | problems - difference - tho - stuff - guy | 45 | 2160_problems_difference_tho_stuff |
| 2161 | rate - likely - minutes - year - chance | 45 | 2161_rate_likely_minutes_year |
| 2162 | set - break - kind - article - crazy | 45 | 2162_set_break_kind_article |
| 2163 | legit - kind - friend - little - real | 45 | 2163_legit_kind_friend_little |
| 2164 | special - fun - sense - interesting - shit | 45 | 2164_special_fun_sense_interesting |
| 2165 | unique - ai - private - fees - game | 45 | 2165_unique_ai_private_fees |
| 2166 | reply - dude - bit - opinion - fun | 45 | 2166_reply_dude_bit_opinion |
| 2167 | fee - services - cryptocurrency - extra - happy | 45 | 2167_fee_services_cryptocurrency_extra |
| 2168 | interested - chat - investment - - | 45 | 2168_interested_chat_investment_ |
| 2169 | worth - trade - week - thanks - bitcoin | 44 | 2169_worth_trade_week_thanks |
| 2170 | personal - sorry - fact - awesome - people | 44 | 2170_personal_sorry_fact_awesome |
| 2171 | comment - dude - friend - solution - awesome | 44 | 2171_comment_dude_friend_solution |
| 2172 | financial - money - order - visit - set | 44 | 2172_financial_money_order_visit |
| 2173 | protocol - ama - excited - ai - spot | 44 | 2173_protocol_ama_excited_ai |
| 2174 | app - code - admin - earn - face | 44 | 2174_app_code_admin_earn |
| 2175 | exciting - ecosystem - announcement - smart - chance | 44 | 2175_exciting_ecosystem_announcement_smart |
| 2176 | current - major - price - btc - multiple | 44 | 2176_current_major_price_btc |
| 2177 | binance - accounts - funds - trade - ok | 44 | 2177_binance_accounts_funds_trade |
| 2178 | alot - contact - perfect - hard - marketing | 44 | 2178_alot_contact_perfect_hard |
| 2179 | missed - search - specific - announcement - situation | 44 | 2179_missed_search_specific_announcement |
| 2180 | liquidity - fix - specific - pool - add | 44 | 2180_liquidity_fix_specific_pool |
| 2181 | previous - attention - massive - buying - event | 44 | 2181_previous_attention_massive_buying |
| 2182 | fun - marketing - ethereum - date - space | 44 | 2182_fun_marketing_ethereum_date |
| 2183 | attention - rest - hour - guide - services | 44 | 2183_attention_rest_hour_guide |
| 2184 | deal - transaction - business - reply - accounts | 44 | 2184_deal_transaction_business_reply |
| 2185 | tg - development - months - supply - liquidity | 44 | 2185_tg_development_months_supply |
| 2186 | coin - near - specific - buying - trust | 44 | 2186_coin_near_specific_buying |
| 2187 | private - huge - marketing - contract - app | 44 | 2187_private_huge_marketing_contract |
| 2188 | contact - quick - high - free - earning | 44 | 2188_contact_quick_high_free |
| 2189 | links - link - mind - dm - post | 44 | 2189_links_link_mind_dm |
| 2190 | success - head - key - successful - likely | 44 | 2190_success_head_key_successful |
| 2191 | defi - product - smart - word - transactions | 44 | 2191_defi_product_smart_word |
| 2192 | dev - shit - announcement - biggest - listing | 44 | 2192_dev_shit_announcement_biggest |
| 2193 | morning - today - good - specific - yesterday | 44 | 2193_morning_today_good_specific |
| 2194 | rest - buddy - glad - dude - nice | 44 | 2194_rest_buddy_glad_dude |
| 2195 | longer - normal - comment - difference - reason | 44 | 2195_longer_normal_comment_difference |
| 2196 | care - family - awesome - little - glad | 44 | 2196_care_family_awesome_little |
| 2197 | deal - form - personal - current - order | 44 | 2197_deal_form_personal_current |
| 2198 | form - games - upcoming - months - higher | 44 | 2198_form_games_upcoming_months |
| 2199 | development - crypto - investors - community - special | 44 | 2199_development_crypto_investors_community |
| 2200 | good - dear - thanks - ok - reply | 44 | 2200_good_dear_thanks_ok |
| 2201 | pls - ethereum - correct - eth - contract | 44 | 2201_pls_ethereum_correct_eth |
| 2202 | research - liquidity - clear - related - process | 44 | 2202_research_liquidity_clear_related |
| 2203 | bot - click - safe - support - official | 44 | 2203_bot_click_safe_support |
| 2204 | hand - service - services - likely - certain | 44 | 2204_hand_service_services_likely |
| 2205 | message - sir - extra - home - available | 44 | 2205_message_sir_extra_home |
| 2206 | bro - service - unique - words - situation | 44 | 2206_bro_service_unique_words |
| 2207 | event - buddy - dude - fun - yesterday | 44 | 2207_event_buddy_dude_fun |
| 2208 | holders - position - coin - new - price | 44 | 2208_holders_position_coin_new |
| 2209 | similar - sense - man - wrong - comment | 44 | 2209_similar_sense_man_wrong |
| 2210 | interesting - moment - important - thanks - sure | 44 | 2210_interesting_moment_important_thanks |
| 2211 | glad - weeks - rest - kind - play | 44 | 2211_glad_weeks_rest_kind |
| 2212 | level - home - deposit - daily - online | 44 | 2212_level_home_deposit_daily |
| 2213 | protocol - technology - earlier - post - special | 44 | 2213_protocol_technology_earlier_post |
| 2214 | guide - follow - example - form - issues | 44 | 2214_guide_follow_example_form |
| 2215 | research - looks - times - buddy - yesterday | 44 | 2215_research_looks_times_buddy |
| 2216 | earlier - early - success - situation - strong | 44 | 2216_earlier_early_success_situation |
| 2217 | option - answer - difference - previous - release | 44 | 2217_option_answer_difference_previous |
| 2218 | clear - specific - article - reply - post | 44 | 2218_clear_specific_article_reply |
| 2219 | ecosystem - shit - dude - buddy - past | 44 | 2219_ecosystem_shit_dude_buddy |
| 2220 | games - unique - updates - plan - amazing | 44 | 2220_games_unique_updates_plan |
| 2221 | awesome - simple - ready - happy - | 44 | 2221_awesome_simple_ready_happy |
| 2222 | clear - focus - true - similar - value | 44 | 2222_clear_focus_true_similar |
| 2223 | space - number - version - features - network | 44 | 2223_space_number_version_features |
| 2224 | turn - step - action - financial - today | 44 | 2224_turn_step_action_financial |
| 2225 | hard - person - tho - true - ones | 43 | 2225_hard_person_tho_true |
| 2226 | lot - - - - | 43 | 2226_lot___ |
| 2227 | risk - try - payment - work - interested | 43 | 2227_risk_try_payment_work |
| 2228 | questions - event - chat - ama - pool | 43 | 2228_questions_event_chat_ama |
| 2229 | block - hour - similar - specific - update | 43 | 2229_block_hour_similar_specific |
| 2230 | transactions - private - secure - data - fact | 43 | 2230_transactions_private_secure_data |
| 2231 | success - things - fact - successful - guys | 43 | 2231_success_things_fact_successful |
| 2232 | face - ur - normal - thought - biggest | 43 | 2232_face_ur_normal_thought |
| 2233 | ethereum - process - trade - bitcoin - link | 43 | 2233_ethereum_process_trade_bitcoin |
| 2234 | fact - cool - job - friend - dear | 43 | 2234_fact_cool_job_friend |
| 2235 | minutes - term - withdrawal - minimum - fee | 43 | 2235_minutes_term_withdrawal_minimum |
| 2236 | word - country - sir - hard - risk | 43 | 2236_word_country_sir_hard |
| 2237 | global - year - risk - capital - high | 43 | 2237_global_year_risk_capital |
| 2238 | claim - connect - listing - binance - early | 43 | 2238_claim_connect_listing_binance |
| 2239 | online - money - life - investment - platform | 43 | 2239_online_money_life_investment |
| 2240 | brother - type - product - fact - little | 43 | 2240_brother_type_product_fact |
| 2241 | head - country - weeks - plan - correct | 43 | 2241_head_country_weeks_plan |
| 2242 | daily - reward - minimum - team - invest | 43 | 2242_daily_reward_minimum_team |
| 2243 | legit - fake - power - opportunity - happy | 43 | 2243_legit_fake_power_opportunity |
| 2244 | chart - previous - dev - global - marketing | 43 | 2244_chart_previous_dev_global |
| 2245 | ppl - dude - answer - person - mate | 43 | 2245_ppl_dude_answer_person |
| 2246 | questions - bot - welcome - official - help | 43 | 2246_questions_bot_welcome_official |
| 2247 | chain - hour - big - hope - trade | 43 | 2247_chain_hour_big_hope |
| 2248 | certain - situation - term - person - matter | 43 | 2248_certain_situation_term_person |
| 2249 | better - fun - things - situation - glad | 43 | 2249_better_fun_things_situation |
| 2250 | help - guy - kind - dude - personal | 43 | 2250_help_guy_kind_dude |
| 2251 | damn - problems - public - report - data | 43 | 2251_damn_problems_public_report |
| 2252 | biggest - run - hand - dude - couple | 43 | 2252_biggest_run_hand_dude |
| 2253 | project - people - - - | 43 | 2253_project_people__ |
| 2254 | minutes - total - wallet - users - correct | 43 | 2254_minutes_total_wallet_users |
| 2255 | supply - total - network - research - user | 43 | 2255_supply_total_network_research |
| 2256 | fast - daily - free - - | 43 | 2256_fast_daily_free_ |
| 2257 | connect - feature - option - wallet - trust | 43 | 2257_connect_feature_option_wallet |
| 2258 | tokens - token - half - example - search | 43 | 2258_tokens_token_half_example |
| 2259 | development - release - interesting - digital - web | 43 | 2259_development_release_interesting_digital |
| 2260 | code - kinda - job - product - dev | 43 | 2260_code_kinda_job_product |
| 2261 | brother - news - buddy - interesting - option | 43 | 2261_brother_news_buddy_interesting |
| 2262 | life - easy - hard - problems - point | 43 | 2262_life_easy_hard_problems |
| 2263 | dude - right - man - brother - fact | 43 | 2263_dude_right_man_brother |
| 2264 | happy - good - day - morning - | 43 | 2264_happy_good_day_morning |
| 2265 | course - head - run - biggest - words | 43 | 2265_course_head_run_biggest |
| 2266 | account - sir - reason - accounts - user | 43 | 2266_account_sir_reason_accounts |
| 2267 | reason - user - message - - | 43 | 2267_reason_user_message_ |
| 2268 | cause - bad - brother - glad - answer | 43 | 2268_cause_bad_brother_glad |
| 2269 | wonder - certain - sense - control - point | 43 | 2269_wonder_certain_sense_control |
| 2270 | chart - market - sell - normal - point | 43 | 2270_chart_market_sell_normal |
| 2271 | pepe - fake - massive - biggest - sure | 43 | 2271_pepe_fake_massive_biggest |
| 2272 | brother - late - guy - night - country | 43 | 2272_brother_late_guy_night |
| 2273 | level - friend - friends - minimum - period | 43 | 2273_level_friend_friends_minimum |
| 2274 | update - version - normal - previous - updates | 43 | 2274_update_version_normal_previous |
| 2275 | volume - rest - price - run - half | 43 | 2275_volume_rest_price_run |
| 2276 | number - ready - - - | 43 | 2276_number_ready__ |
| 2277 | likely - moment - strong - position - brother | 43 | 2277_likely_moment_strong_position |
| 2278 | matter - super - luck - weeks - moment | 43 | 2278_matter_super_luck_weeks |
| 2279 | chart - position - trade - market - buy | 43 | 2279_chart_position_trade_market |
| 2280 | protocol - risk - important - feature - assets | 43 | 2280_protocol_risk_important_feature |
| 2281 | win - fun - biggest - turn - pool | 43 | 2281_win_fun_biggest_turn |
| 2282 | discord - chance - key - public - mate | 43 | 2282_discord_chance_key_public |
| 2283 | bad - today - investment - company - | 43 | 2283_bad_today_investment_company |
| 2284 | home - fun - night - interested - site | 43 | 2284_home_fun_night_interested |
| 2285 | chat - real - person - old - man | 43 | 2285_chat_real_person_old |
| 2286 | minimum - deposit - withdraw - plans - days | 43 | 2286_minimum_deposit_withdraw_plans |
| 2287 | game - links - official - social - website | 43 | 2287_game_links_official_social |
| 2288 | public - taking - experience - chat - page | 42 | 2288_public_taking_experience_chat |
| 2289 | control - damn - problems - head - normal | 42 | 2289_control_damn_problems_head |
| 2290 | job - certain - sense - situation - control | 42 | 2290_job_certain_sense_situation |
| 2291 | ama - win - questions - tomorrow - pm | 42 | 2291_ama_win_questions_tomorrow |
| 2292 | links - new - project - binance - guy | 42 | 2292_links_new_project_binance |
| 2293 | sorry - dude - okay - opinion - mate | 42 | 2293_sorry_dude_okay_opinion |
| 2294 | cool - reply - biggest - crazy - lol | 42 | 2294_cool_reply_biggest_crazy |
| 2295 | short - note - small - people - current | 42 | 2295_short_note_small_people |
| 2296 | face - correct - word - wrong - space | 42 | 2296_face_correct_word_wrong |
| 2297 | risk - level - official - daily - | 42 | 2297_risk_level_official_daily |
| 2298 | amazing - - - - | 42 | 2298_amazing___ |
| 2299 | words - clear - things - yesterday - word | 42 | 2299_words_clear_things_yesterday |
| 2300 | message - dear - limited - fun - dm | 42 | 2300_message_dear_limited_fun |
| 2301 | glad - welcome - group - telegram - best | 42 | 2301_glad_welcome_group_telegram |
| 2302 | ai - space - end - year - smart | 42 | 2302_ai_space_end_year |
| 2303 | spot - chain - trade - tokens - week | 42 | 2303_spot_chain_trade_tokens |
| 2304 | bro - specific - previous - guys - ones | 42 | 2304_bro_specific_previous_guys |
| 2305 | claim - luck - holders - chance - reward | 42 | 2305_claim_luck_holders_chance |
| 2306 | withdraw - withdrawal - working - web - issue | 42 | 2306_withdraw_withdrawal_working_web |
| 2307 | data - focus - rules - community - discord | 42 | 2307_data_focus_rules_community |
| 2308 | code - worth - app - exchange - email | 42 | 2308_code_worth_app_exchange |
| 2309 | link - - - - | 42 | 2309_link___ |
| 2310 | pm - payment - - - | 42 | 2310_pm_payment__ |
| 2311 | wonder - product - clear - ones - lot | 42 | 2311_wonder_product_clear_ones |
| 2312 | product - brother - normal - clear - tho | 42 | 2312_product_brother_normal_clear |
| 2313 | massive - ama - marketing - holders - transaction | 42 | 2313_massive_ama_marketing_holders |
| 2314 | defi - assets - simple - high - users | 42 | 2314_defi_assets_simple_high |
| 2315 | example - binance - code - claim - reward | 42 | 2315_example_binance_code_claim |
| 2316 | value - assets - worth - drop - example | 42 | 2316_value_assets_worth_drop |
| 2317 | period - minutes - binance - protocol - profit | 42 | 2317_period_minutes_binance_protocol |
| 2318 | real - kind - drop - second - awesome | 42 | 2318_real_kind_drop_second |
| 2319 | face - look - shit - lol - wrong | 42 | 2319_face_look_shit_lol |
| 2320 | major - chance - awesome - yesterday - dev | 42 | 2320_major_chance_awesome_yesterday |
| 2321 | point - excited - positive - hope - little | 42 | 2321_point_excited_positive_hope |
| 2322 | correct - trading - trust - reply - late | 42 | 2322_correct_trading_trust_reply |
| 2323 | contact - words - short - successful - profit | 42 | 2323_contact_words_short_successful |
| 2324 | coin - trust - chain - coins - search | 42 | 2324_coin_trust_chain_coins |
| 2325 | mate - hand - difference - major - building | 42 | 2325_mate_hand_difference_major |
| 2326 | profit - social - earning - real - sure | 42 | 2326_profit_social_earning_real |
| 2327 | holders - kind - project - token - right | 42 | 2327_holders_kind_project_token |
| 2328 | guy - person - old - case - normal | 42 | 2328_guy_person_old_case |
| 2329 | welcome - official - group - couple - coin | 42 | 2329_welcome_official_group_couple |
| 2330 | specific - mind - biggest - shit - fine | 42 | 2330_specific_mind_biggest_shit |
| 2331 | run - hand - head - guy - special | 42 | 2331_run_hand_head_guy |
| 2332 | head - yesterday - stuff - guy - type | 42 | 2332_head_yesterday_stuff_guy |
| 2333 | term - long - short - tho - thinking | 42 | 2333_term_long_short_tho |
| 2334 | fair - list - listing - projects - social | 42 | 2334_fair_list_listing_projects |
| 2335 | report - latest - year - period - sign | 42 | 2335_report_latest_year_period |
| 2336 | similar - page - day - dm - | 42 | 2336_similar_page_day_dm |
| 2337 | awesome - question - buying - dev - shit | 42 | 2337_awesome_question_buying_dev |
| 2338 | binance - details - simple - wallet - question | 42 | 2338_binance_details_simple_wallet |
| 2339 | protocol - funds - team - - | 42 | 2339_protocol_funds_team_ |
| 2340 | turn - group - chat - value - official | 42 | 2340_turn_group_chat_value |
| 2341 | feature - recent - utc - excited - sign | 41 | 2341_feature_recent_utc_excited |
| 2342 | latest - data - version - issues - build | 41 | 2342_latest_data_version_issues |
| 2343 | ecosystem - funds - great - people - | 41 | 2343_ecosystem_funds_great_people |
| 2344 | thing - certain - head - situation - reason | 41 | 2344_thing_certain_head_situation |
| 2345 | end - point - person - link - level | 41 | 2345_end_point_person_link |
| 2346 | ways - bot - potential - future - post | 41 | 2346_ways_bot_potential_future |
| 2347 | personal - information - user - attention - asking | 41 | 2347_personal_information_user_attention |
| 2348 | admin - look - guy - ones - group | 41 | 2348_admin_look_guy_ones |
| 2349 | spot - real - page - guys - channel | 41 | 2349_spot_real_page_guys |
| 2350 | rest - contact - marketing - fast - great | 41 | 2350_rest_contact_marketing_fast |
| 2351 | hours - new - product - hard - real | 41 | 2351_hours_new_product_hard |
| 2352 | solution - assets - process - network - bitcoin | 41 | 2352_solution_assets_process_network |
| 2353 | attention - person - interesting - stuff - type | 41 | 2353_attention_person_interesting_stuff |
| 2354 | list - fake - admin - information - official | 41 | 2354_list_fake_admin_information |
| 2355 | fix - showing - network - contract - ethereum | 41 | 2355_fix_showing_network_contract |
| 2356 | global - business - trading - profit - investors | 41 | 2356_global_business_trading_profit |
| 2357 | couple - reply - thanks - question - minutes | 41 | 2357_couple_reply_thanks_question |
| 2358 | break - major - support - price - short | 41 | 2358_break_major_support_price |
| 2359 | holders - ethereum - supply - situation - recent | 41 | 2359_holders_ethereum_supply_situation |
| 2360 | transaction - note - trading - success - trade | 41 | 2360_transaction_note_trading_success |
| 2361 | step - post - contact - work - life | 41 | 2361_step_post_contact_work |
| 2362 | brother - cool - nice - sense - tho | 41 | 2362_brother_cool_nice_sense |
| 2363 | tho - things - face - awesome - man | 41 | 2363_tho_things_face_awesome |
| 2364 | specific - similar - issues - sense - article | 41 | 2364_specific_similar_issues_sense |
| 2365 | fake - accounts - account - site - note | 41 | 2365_fake_accounts_account_site |
| 2366 | write - live - term - high - interested | 41 | 2366_write_live_term_high |
| 2367 | control - key - chart - bitcoin - network | 41 | 2367_control_key_chart_bitcoin |
| 2368 | available - note - phone - number - hand | 41 | 2368_available_note_phone_number |
| 2369 | run - value - bitcoin - new - spot | 41 | 2369_run_value_bitcoin_new |
| 2370 | holders - space - social - big - assets | 41 | 2370_holders_space_social_big |
| 2371 | longer - reply - update - moment - bit | 41 | 2371_longer_reply_update_moment |
| 2372 | liquidity - smart - security - chain - price | 41 | 2372_liquidity_smart_security_chain |
| 2373 | positive - couple - exciting - current - holders | 41 | 2373_positive_couple_exciting_current |
| 2374 | message - available - - - | 41 | 2374_message_available__ |
| 2375 | difference - trading - specific - trade - experience | 41 | 2375_difference_trading_specific_trade |
| 2376 | situation - kinda - listing - simple - announcement | 41 | 2376_situation_kinda_listing_simple |
| 2377 | strong - major - stuff - true - shit | 41 | 2377_strong_major_stuff_true |
| 2378 | previous - media - fee - marketing - smart | 41 | 2378_previous_media_fee_marketing |
| 2379 | legit - interested - little - real - money | 41 | 2379_legit_interested_little_real |
| 2380 | bot - ways - interested - simple - way | 41 | 2380_bot_ways_interested_simple |
| 2381 | likely - lots - kind - position - stop | 41 | 2381_likely_lots_kind_position |
| 2382 | feature - key - service - users - new | 41 | 2382_feature_key_service_users |
| 2383 | friends - official - profit - group - active | 41 | 2383_friends_official_profit_group |
| 2384 | half - sorry - ways - close - weeks | 41 | 2384_half_sorry_ways_close |
| 2385 | digital - biggest - simple - choose - members | 41 | 2385_digital_biggest_simple_choose |
| 2386 | financial - ethereum - coins - minutes - crypto | 41 | 2386_financial_ethereum_coins_minutes |
| 2387 | ones - low - - - | 41 | 2387_ones_low__ |
| 2388 | feature - awesome - glad - spot - site | 41 | 2388_feature_awesome_glad_spot |
| 2389 | period - thanks - days - weeks - issues | 41 | 2389_period_thanks_days_weeks |
| 2390 | mate - head - reply - cause - buddy | 41 | 2390_mate_head_reply_cause |
| 2391 | sense - version - old - cool - matter | 41 | 2391_sense_version_old_cool |
| 2392 | early - rewards - ai - access - power | 41 | 2392_early_rewards_ai_access |
| 2393 | possible - ways - sorry - sure - idea | 41 | 2393_possible_ways_sorry_sure |
| 2394 | blockchain - ama - projects - technology - members | 41 | 2394_blockchain_ama_projects_technology |
| 2395 | games - blockchain - development - future - tg | 41 | 2395_games_blockchain_development_future |
| 2396 | couple - fake - fine - idea - case | 41 | 2396_couple_fake_fine_idea |
| 2397 | hand - opinion - normal - ones - country | 41 | 2397_hand_opinion_normal_ones |
| 2398 | example - stuff - option - site - kind | 41 | 2398_example_stuff_option_site |
| 2399 | drop - old - funds - deal - hour | 41 | 2399_drop_old_funds_deal |
| 2400 | sign - key - address - multiple - feature | 41 | 2400_sign_key_address_multiple |
| 2401 | pool - minimum - step - deposit - action | 41 | 2401_pool_minimum_step_deposit |
| 2402 | active - - - - | 41 | 2402_active___ |
| 2403 | admin - details - personal - official - bot | 40 | 2403_admin_details_personal_official |
| 2404 | nft - playing - games - features - space | 40 | 2404_nft_playing_games_features |
| 2405 | months - nft - btc - help - day | 40 | 2405_months_nft_btc_help |
| 2406 | mate - shit - idea - bro - happy | 40 | 2406_mate_shit_idea_bro |
| 2407 | tho - buddy - supply - moment - bit | 40 | 2407_tho_buddy_supply_moment |
| 2408 | group - person - dude - brother - normal | 40 | 2408_group_person_dude_brother |
| 2409 | claim - network - tokens - address - chain | 40 | 2409_claim_network_tokens_address |
| 2410 | normal - difference - half - country - bad | 40 | 2410_normal_difference_half_country |
| 2411 | ama - product - features - development - latest | 40 | 2411_ama_product_features_development |
| 2412 | wallet - half - secure - key - point | 40 | 2412_wallet_half_secure_key |
| 2413 | short - ready - best - platform - | 40 | 2413_short_ready_best_platform |
| 2414 | minimum - market - price - hours - deposit | 40 | 2414_minimum_market_price_hours |
| 2415 | link - - - - | 40 | 2415_link___ |
| 2416 | success - successful - single - small - certain | 40 | 2416_success_successful_single_small |
| 2417 | - - - - | 40 | 2417____ |
| 2418 | step - words - opinion - real - things | 40 | 2418_step_words_opinion_real |
| 2419 | rest - dude - fine - awesome - glad | 40 | 2419_rest_dude_fine_awesome |
| 2420 | building - lots - couple - different - moment | 40 | 2420_building_lots_couple_different |
| 2421 | multiple - hour - info - services - channel | 40 | 2421_multiple_hour_info_services |
| 2422 | liquidity - period - contract - fix - awesome | 40 | 2422_liquidity_period_contract_fix |
| 2423 | user - single - change - clear - major | 40 | 2423_user_single_change_clear |
| 2424 | global - worth - minimum - form - extra | 40 | 2424_global_worth_minimum_form |
| 2425 | order - report - guy - video - control | 40 | 2425_order_report_guy_video |
| 2426 | projects - guy - week - crazy - guys | 40 | 2426_projects_guy_week_crazy |
| 2427 | issues - issue - problem - yesterday - web | 40 | 2427_issues_issue_problem_yesterday |
| 2428 | total - amazing - day - payment - biggest | 40 | 2428_total_amazing_day_payment |
| 2429 | great - today - best - good - works | 40 | 2429_great_today_best_good |
| 2430 | comment - cause - face - shit - half | 40 | 2430_comment_cause_face_shit |
| 2431 | days - sense - day - dude - difference | 40 | 2431_days_sense_day_dude |
| 2432 | online - interested - job - home - pay | 40 | 2432_online_interested_job_home |
| 2433 | liquidity - marketing - rewards - token - best | 40 | 2433_liquidity_marketing_rewards_token |
| 2434 | awesome - simple - open - easy - available | 40 | 2434_awesome_simple_open_easy |
| 2435 | email - course - business - service - phone | 40 | 2435_email_course_business_service |
| 2436 | ll - dude - wrong - shit - bit | 40 | 2436_ll_dude_wrong_shit |
| 2437 | dm - - - - | 40 | 2437_dm___ |
| 2438 | option - game - technology - kind - idea | 40 | 2438_option_game_technology_kind |
| 2439 | situation - friend - word - nice - family | 40 | 2439_situation_friend_word_nice |
| 2440 | past - moment - alot - course - morning | 40 | 2440_past_moment_alot_course |
| 2441 | run - damn - guy - sense - tho | 40 | 2441_run_damn_guy_sense |
| 2442 | date - guys - today - better - | 40 | 2442_date_guys_today_better |
| 2443 | - - - - | 40 | 2443____ |
| 2444 | short - control - fact - tho - ones | 40 | 2444_short_control_fact_tho |
| 2445 | welcome - potential - glad - guys - community | 40 | 2445_welcome_potential_glad_guys |
| 2446 | event - type - month - mate - luck | 40 | 2446_event_type_month_mate |
| 2447 | head - release - holders - solution - recent | 40 | 2447_head_release_holders_solution |
| 2448 | fee - sir - protocol - connect - luck | 40 | 2448_fee_sir_protocol_connect |
| 2449 | available - phone - real - payment - fast | 40 | 2449_available_phone_real_payment |
| 2450 | list - tho - yesterday - past - moment | 40 | 2450_list_tho_yesterday_past |
| 2451 | morning - happy - good - nice - fun | 40 | 2451_morning_happy_good_nice |
| 2452 | problems - example - perfect - home - job | 40 | 2452_problems_example_perfect_home |
| 2453 | answer - simple - question - guy - lol | 40 | 2453_answer_simple_question_guy |
| 2454 | questions - difference - similar - rate - things | 40 | 2454_questions_difference_similar_rate |
| 2455 | trade - short - deposit - - | 40 | 2455_trade_short_deposit_ |
| 2456 | talk - things - people - - | 40 | 2456_talk_things_people_ |
| 2457 | alot - care - amazing - active - withdrawal | 40 | 2457_alot_care_amazing_active |
| 2458 | situation - fun - love - attention - kind | 40 | 2458_situation_fun_love_attention |
| 2459 | damn - money - sense - life - lots | 40 | 2459_damn_money_sense_life |
| 2460 | order - questions - number - email - video | 40 | 2460_order_questions_number_email |
| 2461 | reason - links - problem - man - info | 40 | 2461_reason_links_problem_man |
| 2462 | swap - ethereum - point - eth - previous | 40 | 2462_swap_ethereum_point_eth |
| 2463 | pm - time - questions - service - plan | 40 | 2463_pm_time_questions_service |
| 2464 | brother - situation - product - major - buddy | 40 | 2464_brother_situation_product_major |
| 2465 | upcoming - ai - ways - new - use | 40 | 2465_upcoming_ai_ways_new |
| 2466 | research - features - token - similar - kind | 40 | 2466_research_features_token_similar |
| 2467 | family - chat - case - questions - form | 40 | 2467_family_chat_case_questions |
| 2468 | half - run - building - ecosystem - focus | 40 | 2468_half_run_building_ecosystem |
| 2469 | binance - alot - recent - release - moment | 40 | 2469_binance_alot_recent_release |
| 2470 | risk - level - official - daily - | 40 | 2470_risk_level_official_daily |
| 2471 | plans - trade - spot - link - investors | 40 | 2471_plans_trade_spot_link |
| 2472 | ai - technology - space - risk - crypto | 40 | 2472_ai_technology_space_risk |
| 2473 | night - step - good - tomorrow - hard | 39 | 2473_night_step_good_tomorrow |
| 2474 | network - site - block - problems - address | 39 | 2474_network_site_block_problems |
| 2475 | report - week - market - bitcoin - new | 39 | 2475_report_week_market_bitcoin |
| 2476 | telegram - friends - good - real - | 39 | 2476_telegram_friends_good_real |
| 2477 | interested - interesting - normal - recent - question | 39 | 2477_interested_interesting_normal_recent |
| 2478 | control - old - plus - type - different | 39 | 2478_control_old_plus_type |
| 2479 | discord - development - site - exciting - website | 39 | 2479_discord_development_site_exciting |
| 2480 | chain - similar - sir - special - supply | 39 | 2480_chain_similar_sir_special |
| 2481 | ai - financial - article - hand - global | 39 | 2481_ai_financial_article_hand |
| 2482 | stuff - type - games - official - case | 39 | 2482_stuff_type_games_official |
| 2483 | protocol - level - best - - | 39 | 2483_protocol_level_best_ |
| 2484 | hours - new - people - money - | 39 | 2484_hours_new_people_money |
| 2485 | - - - - | 39 | 2485____ |
| 2486 | reply - answer - late - yesterday - positive | 39 | 2486_reply_answer_late_yesterday |
| 2487 | listing - contact - business - development - exchange | 39 | 2487_listing_contact_business_development |
| 2488 | chat - talk - hand - dude - question | 39 | 2488_chat_talk_hand_dude |
| 2489 | hope - man - mate - true - guys | 39 | 2489_hope_man_mate_true |
| 2490 | financial - technology - users - potential - guy | 39 | 2490_financial_technology_users_potential |
| 2491 | control - successful - best - trading - order | 39 | 2491_control_successful_best_trading |
| 2492 | multiple - main - crazy - yesterday - weeks | 39 | 2492_multiple_main_crazy_yesterday |
| 2493 | bot - looks - late - luck - guy | 39 | 2493_bot_looks_late_luck |
| 2494 | couple - biggest - research - early - home | 39 | 2494_couple_biggest_research_early |
| 2495 | investors - bitcoin - investment - company - time | 39 | 2495_investors_bitcoin_investment_company |
| 2496 | dude - sense - latest - article - guys | 39 | 2496_dude_sense_latest_article |
| 2497 | way - new - services - ready - experience | 39 | 2497_way_new_services_ready |
| 2498 | wallet - company - - - | 39 | 2498_wallet_company__ |
| 2499 | night - damn - head - shit - option | 39 | 2499_night_damn_head_shit |
| 2500 | future - life - community - today - times | 39 | 2500_future_life_community_today |
| 2501 | total - awesome - perfect - nice - post | 39 | 2501_total_awesome_perfect_nice |
| 2502 | brother - money - work - way - ways | 39 | 2502_brother_money_work_way |
| 2503 | issues - financial - help - extra - support | 39 | 2503_issues_financial_help_extra |
| 2504 | app - works - phone - problem - feature | 39 | 2504_app_works_phone_problem |
| 2505 | early - late - tbh - glad - crazy | 39 | 2505_early_late_tbh_glad |
| 2506 | smart - power - normal - features - minutes | 39 | 2506_smart_power_normal_features |
| 2507 | fine - dude - brother - sorry - tho | 39 | 2507_fine_dude_brother_sorry |
| 2508 | hours - service - similar - video - site | 39 | 2508_hours_service_similar_video |
| 2509 | discord - follow - date - social - latest | 39 | 2509_discord_follow_date_social |
| 2510 | official - difference - luck - ones - word | 39 | 2510_official_difference_luck_ones |
| 2511 | certain - moment - simple - man - cool | 39 | 2511_certain_moment_simple_man |
| 2512 | page - active - details - special - pm | 39 | 2512_page_active_details_special |
| 2513 | past - works - plan - process - launch | 39 | 2513_past_works_plan_process |
| 2514 | report - links - private - admin - support | 39 | 2514_report_links_private_admin |
| 2515 | ecosystem - situation - focus - smart - projects | 39 | 2515_ecosystem_situation_focus_smart |
| 2516 | comment - guys - active - man - sure | 39 | 2516_comment_guys_active_man |
| 2517 | utc - link - man - wallet - | 39 | 2517_utc_link_man_wallet |
| 2518 | example - point - shit - ones - early | 39 | 2518_example_point_shit_ones |
| 2519 | game - video - guide - dear - new | 39 | 2519_game_video_guide_dear |
| 2520 | dude - sense - difference - major - nice | 39 | 2520_dude_sense_difference_major |
| 2521 | current - free - bot - information - certain | 39 | 2521_current_free_bot_information |
| 2522 | price - rate - fun - point - fact | 39 | 2522_price_rate_fun_point |
| 2523 | multiple - perfect - high - real - payment | 39 | 2523_multiple_perfect_high_real |
| 2524 | clear - weeks - shit - word - simple | 39 | 2524_clear_weeks_shit_word |
| 2525 | complete - token - wallet - join - rewards | 39 | 2525_complete_token_wallet_join |
| 2526 | late - love - matter - job - better | 39 | 2526_late_love_matter_job |
| 2527 | site - deposit - sign - possible - daily | 39 | 2527_site_deposit_sign_possible |
| 2528 | technology - game - blockchain - value - turn | 39 | 2528_technology_game_blockchain_value |
| 2529 | digital - rules - issues - assets - certain | 39 | 2529_digital_rules_issues_assets |
| 2530 | rate - difference - hand - certain - situation | 39 | 2530_rate_difference_hand_certain |
| 2531 | hour - right - coins - transactions - eth | 39 | 2531_hour_right_coins_transactions |
| 2532 | development - simple - site - minutes - special | 39 | 2532_development_simple_site_minutes |
| 2533 | end - event - announcement - difference - discord | 39 | 2533_end_event_announcement_difference |
| 2534 | problems - feature - tho - yesterday - better | 39 | 2534_problems_feature_tho_yesterday |
| 2535 | free - opinion - situation - ways - past | 39 | 2535_free_opinion_situation_ways |
| 2536 | similar - solution - period - hour - minutes | 39 | 2536_similar_solution_period_hour |
| 2537 | stop - key - success - example - order | 38 | 2537_stop_key_success_example |
| 2538 | support - - - - | 38 | 2538_support___ |
| 2539 | wonder - similar - difference - product - crazy | 38 | 2539_wonder_similar_difference_product |
| 2540 | ethereum - things - blockchain - article - post | 38 | 2540_ethereum_things_blockchain_article |
| 2541 | answer - night - questions - ready - question | 38 | 2541_answer_night_questions_ready |
| 2542 | research - amazing - people - days - good | 38 | 2542_research_amazing_people_days |
| 2543 | opinion - past - earlier - things - mean | 38 | 2543_opinion_past_earlier_things |
| 2544 | free - minimum - bot - welcome - hours | 38 | 2544_free_minimum_bot_welcome |
| 2545 | cool - minutes - app - cryptocurrency - extra | 38 | 2545_cool_minutes_app_cryptocurrency |
| 2546 | level - successful - risk - info - work | 38 | 2546_level_successful_risk_info |
| 2547 | second - trading - position - weeks - higher | 38 | 2547_second_trading_position_weeks |
| 2548 | unique - space - different - potential - claim | 38 | 2548_unique_space_different_potential |
| 2549 | place - sir - month - thanks - profit | 38 | 2549_place_sir_month_thanks |
| 2550 | swap - solution - feature - transaction - example | 38 | 2550_swap_solution_feature_transaction |
| 2551 | tho - issues - certain - rate - friend | 38 | 2551_tho_issues_certain_rate |
| 2552 | love - buddy - problems - special - past | 38 | 2552_love_buddy_problems_special |
| 2553 | fun - minutes - position - note - run | 38 | 2553_fun_minutes_position_note |
| 2554 | site - earning - minimum - sign - early | 38 | 2554_site_earning_minimum_sign |
| 2555 | guy - comment - clear - pretty - bit | 38 | 2555_guy_comment_clear_pretty |
| 2556 | rules - transactions - worth - minimum - claim | 38 | 2556_rules_transactions_worth_minimum |
| 2557 | excited - home - care - family - contact | 38 | 2557_excited_home_care_family |
| 2558 | current - value - market - cryptocurrency - global | 38 | 2558_current_value_market_cryptocurrency |
| 2559 | taking - tho - luck - alot - night | 38 | 2559_taking_tho_luck_alot |
| 2560 | sign - positive - similar - opinion - good | 38 | 2560_sign_positive_similar_opinion |
| 2561 | selling - fix - deal - available - order | 38 | 2561_selling_fix_deal_available |
| 2562 | personal - drop - pm - thanks - time | 38 | 2562_personal_drop_pm_thanks |
| 2563 | binance - services - platform - new - private | 38 | 2563_binance_services_platform_new |
| 2564 | hold - mate - holding - opinion - close | 38 | 2564_hold_mate_holding_opinion |
| 2565 | connect - wallet - happy - website - site | 38 | 2565_connect_wallet_happy_website |
| 2566 | telegram - launch - public - utc - current | 38 | 2566_telegram_launch_public_utc |
| 2567 | months - site - accounts - services - key | 38 | 2567_months_site_accounts_services |
| 2568 | rules - network - needs - bitcoin - | 38 | 2568_rules_network_needs_bitcoin |
| 2569 | today - hand - total - old - chat | 38 | 2569_today_hand_total_old |
| 2570 | ready - night - lot - good - | 38 | 2570_ready_night_lot_good |
| 2571 | long - life - key - action - success | 38 | 2571_long_life_key_action |
| 2572 | comment - dev - likely - head - reply | 38 | 2572_comment_dev_likely_head |
| 2573 | games - chart - contract - network - available | 38 | 2573_games_chart_contract_network |
| 2574 | option - second - exchange - times - account | 38 | 2574_option_second_exchange_times |
| 2575 | successful - secure - works - amazing - cryptocurrency | 38 | 2575_successful_secure_works_amazing |
| 2576 | look - capital - investors - amazing - company | 38 | 2576_look_capital_investors_amazing |
| 2577 | technology - unique - mind - app - project | 38 | 2577_technology_unique_mind_app |
| 2578 | kinda - ur - block - situation - product | 38 | 2578_kinda_ur_block_situation |
| 2579 | comment - number - previous - small - alot | 38 | 2579_comment_number_previous_small |
| 2580 | turn - site - step - profit - money | 38 | 2580_turn_site_step_profit |
| 2581 | user - chat - way - info - users | 38 | 2581_user_chat_way_info |
| 2582 | buying - crypto - option - alright - dude | 38 | 2582_buying_crypto_option_alright |
| 2583 | better - case - higher - glad - trust | 38 | 2583_better_case_higher_glad |
| 2584 | awesome - twitter - list - previous - update | 38 | 2584_awesome_twitter_list_previous |
| 2585 | months - weeks - search - article - channel | 38 | 2585_months_weeks_search_article |
| 2586 | public - ones - latest - private - projects | 38 | 2586_public_ones_latest_private |
| 2587 | near - future - twitter - - | 38 | 2587_near_future_twitter_ |
| 2588 | potential - contact - - - | 38 | 2588_potential_contact__ |
| 2589 | - - - - | 38 | 2589____ |
| 2590 | night - mate - correct - wish - fine | 38 | 2590_night_mate_correct_wish |
| 2591 | alot - site - free - video - minimum | 38 | 2591_alot_site_free_video |
| 2592 | comment - twitter - fine - massive - couple | 38 | 2592_comment_twitter_fine_massive |
| 2593 | crypto - use - market - people - team | 38 | 2593_crypto_use_market_people |
| 2594 | half - product - sir - long - place | 38 | 2594_half_product_sir_long |
| 2595 | face - sense - chance - little - moment | 38 | 2595_face_sense_chance_little |
| 2596 | trade - bit - - - | 38 | 2596_trade_bit__ |
| 2597 | right - things - wrong - - | 38 | 2597_right_things_wrong_ |
| 2598 | chart - position - trade - buy - market | 38 | 2598_chart_position_trade_buy |
| 2599 | version - details - dude - different - focus | 38 | 2599_version_details_dude_different |
| 2600 | liquidity - half - stuff - position - difference | 38 | 2600_liquidity_half_stuff_position |
| 2601 | alot - cryptocurrency - means - investing - today | 38 | 2601_alot_cryptocurrency_means_investing |
| 2602 | earlier - awesome - ones - single - sorry | 38 | 2602_earlier_awesome_ones_single |
| 2603 | late - years - secure - profits - service | 38 | 2603_late_years_secure_profits |
| 2604 | glad - awesome - welcome - dude - thanks | 38 | 2604_glad_awesome_welcome_dude |
| 2605 | complete - days - daily - deposit - day | 38 | 2605_complete_days_daily_deposit |
| 2606 | reply - email - job - course - specific | 38 | 2606_reply_email_job_course |
| 2607 | alright - daily - hand - guys - ready | 38 | 2607_alright_daily_hand_guys |
| 2608 | tho - dev - head - space - recent | 38 | 2608_tho_dev_head_space |
| 2609 | form - certain - support - report - fair | 38 | 2609_form_certain_support_report |
| 2610 | wonder - spot - luck - shit - looks | 38 | 2610_wonder_spot_luck_shit |
| 2611 | trading - new - platform - - | 38 | 2611_trading_new_platform_ |
| 2612 | head - comment - face - stuff - shit | 37 | 2612_head_comment_face_stuff |
| 2613 | personal - different - normal - trust - focus | 37 | 2613_personal_different_normal_trust |
| 2614 | previous - launch - huge - contract - coin | 37 | 2614_previous_launch_huge_contract |
| 2615 | features - unique - early - code - ai | 37 | 2615_features_unique_early_code |
| 2616 | head - shit - main - minutes - good | 37 | 2616_head_shit_main_minutes |
| 2617 | earlier - country - date - search - work | 37 | 2617_earlier_country_date_search |
| 2618 | binance - btc - - - | 37 | 2618_binance_btc__ |
| 2619 | blockchain - market - trading - bitcoin - run | 37 | 2619_blockchain_market_trading_bitcoin |
| 2620 | yesterday - profit - couple - plan - short | 37 | 2620_yesterday_profit_couple_plan |
| 2621 | pool - deposit - win - multiple - massive | 37 | 2621_pool_deposit_win_multiple |
| 2622 | fine - sorry - nice - interesting - long | 37 | 2622_fine_sorry_nice_interesting |
| 2623 | year - extra - message - payment - big | 37 | 2623_year_extra_message_payment |
| 2624 | good - hand - mate - dear - man | 37 | 2624_good_hand_mate_dear |
| 2625 | projects - opinion - current - market - list | 37 | 2625_projects_opinion_current_market |
| 2626 | launch - multiple - huge - listing - marketing | 37 | 2626_launch_multiple_huge_listing |
| 2627 | token - half - case - binance - liquidity | 37 | 2627_token_half_case_binance |
| 2628 | sir - fun - - - | 37 | 2628_sir_fun__ |
| 2629 | dear - risk - friends - offer - profits | 37 | 2629_dear_risk_friends_offer |
| 2630 | chain - latest - step - welcome - questions | 37 | 2630_chain_latest_step_welcome |
| 2631 | cool - previous - interesting - research - sign | 37 | 2631_cool_previous_interesting_research |
| 2632 | problems - global - financial - years - experience | 37 | 2632_problems_global_financial_years |
| 2633 | ones - search - answer - number - plans | 37 | 2633_ones_search_answer_number |
| 2634 | coin - coins - hours - great - good | 37 | 2634_coin_coins_hours_great |
| 2635 | biggest - higher - ready - btc - group | 37 | 2635_biggest_higher_ready_btc |
| 2636 | nice - cool - dear - mind - mate | 37 | 2636_nice_cool_dear_mind |
| 2637 | capital - fact - business - luck - guy | 37 | 2637_capital_fact_business_luck |
| 2638 | research - damn - matter - claim - sorry | 37 | 2638_research_damn_matter_claim |
| 2639 | phone - action - week - work - money | 37 | 2639_phone_action_week_work |
| 2640 | year - - - - | 37 | 2640_year___ |
| 2641 | ethereum - special - opinion - volume - fact | 37 | 2641_ethereum_special_opinion_volume |
| 2642 | problems - example - awesome - night - minutes | 37 | 2642_problems_example_awesome_night |
| 2643 | update - special - fact - late - talking | 37 | 2643_update_special_fact_late |
| 2644 | friend - stuff - thanks - words - idea | 37 | 2644_friend_stuff_thanks_words |
| 2645 | dude - bro - thing - crazy - cool | 37 | 2645_dude_bro_thing_crazy |
| 2646 | market - investment - - - | 37 | 2646_market_investment__ |
| 2647 | guys - bad - good - fine - mate | 37 | 2647_guys_bad_good_fine |
| 2648 | rate - profit - trading - market - close | 37 | 2648_rate_profit_trading_market |
| 2649 | swap - usdt - issues - period - transaction | 37 | 2649_swap_usdt_issues_period |
| 2650 | kinda - stuff - type - multiple - phone | 37 | 2650_kinda_stuff_type_multiple |
| 2651 | company - - - - | 37 | 2651_company___ |
| 2652 | certain - fun - previous - things - interesting | 37 | 2652_certain_fun_previous_things |
| 2653 | bad - friend - mate - tho - awesome | 37 | 2653_bad_friend_mate_tho |
| 2654 | second - hour - sir - opinion - yesterday | 37 | 2654_second_hour_sir_opinion |
| 2655 | smart - future - version - simple - article | 37 | 2655_smart_future_version_simple |
| 2656 | assets - cryptocurrency - share - btc - crypto | 37 | 2656_assets_cryptocurrency_share_btc |
| 2657 | normal - hours - luck - yesterday - hour | 37 | 2657_normal_hours_luck_yesterday |
| 2658 | recent - second - real - successful - fake | 37 | 2658_recent_second_real_successful |
| 2659 | important - report - members - admin - ones | 37 | 2659_important_report_members_admin |
| 2660 | safe - reason - admin - chat - financial | 37 | 2660_safe_reason_admin_chat |
| 2661 | trade - supply - minutes - short - total | 37 | 2661_trade_supply_minutes_short |
| 2662 | half - awesome - yesterday - bro - hour | 37 | 2662_half_awesome_yesterday_bro |
| 2663 | transaction - account - deposit - active - app | 37 | 2663_transaction_account_deposit_active |
| 2664 | build - version - date - available - note | 37 | 2664_build_version_date_available |
| 2665 | games - home - better - guy - game | 37 | 2665_games_home_better_guy |
| 2666 | ones - rest - search - ready - start | 37 | 2666_ones_rest_search_ready |
| 2667 | rules - case - report - digital - assets | 37 | 2667_rules_case_report_digital |
| 2668 | public - coins - cryptocurrency - years - bitcoin | 37 | 2668_public_coins_cryptocurrency_years |
| 2669 | protocol - needs - care - cool - chain | 37 | 2669_protocol_needs_care_cool |
| 2670 | defi - week - ethereum - features - wallet | 37 | 2670_defi_week_ethereum_features |
| 2671 | video - daily - term - minutes - deposit | 37 | 2671_video_daily_term_minutes |
| 2672 | fair - unique - development - digital - launch | 37 | 2672_fair_unique_development_digital |
| 2673 | research - recent - cryptocurrency - years - | 37 | 2673_research_recent_cryptocurrency_years |
| 2674 | words - head - article - difference - night | 37 | 2674_words_head_article_difference |
| 2675 | swap - similar - dm - today - follow | 37 | 2675_swap_similar_dm_today |
| 2676 | media - social - limited - - | 37 | 2676_media_social_limited_ |
| 2677 | funds - users - wallet - link - news | 37 | 2677_funds_users_wallet_link |
| 2678 | home - family - hour - couple - break | 37 | 2678_home_family_hour_couple |
| 2679 | ethereum - block - chain - release - recent | 37 | 2679_ethereum_block_chain_release |
| 2680 | binance - visit - video - research - channel | 37 | 2680_binance_visit_video_research |
| 2681 | dear - true - risk - friends - offer | 37 | 2681_dear_true_risk_friends |
| 2682 | buying - solution - buddy - fair - moment | 37 | 2682_buying_solution_buddy_fair |
| 2683 | strong - dev - product - grow - opinion | 37 | 2683_strong_dev_product_grow |
| 2684 | win - comment - special - sign - utc | 37 | 2684_win_comment_special_sign |
| 2685 | biggest - date - turn - money - world | 37 | 2685_biggest_date_turn_money |
| 2686 | risk - level - official - daily - | 37 | 2686_risk_level_official_daily |
| 2687 | list - big - link - today - | 37 | 2687_list_big_link_today |
| 2688 | friend - legit - fee - cryptocurrency - stop | 37 | 2688_friend_legit_fee_cryptocurrency |
| 2689 | ecosystem - today - free - wallet - | 37 | 2689_ecosystem_today_free_wallet |
| 2690 | web - type - sorry - issue - problem | 36 | 2690_web_type_sorry_issue |
| 2691 | chat - announcement - official - problems - comment | 36 | 2691_chat_announcement_official_problems |
| 2692 | block - space - data - shit - cool | 36 | 2692_block_space_data_shit |
| 2693 | sorry - luck - true - bad - better | 36 | 2693_sorry_luck_true_bad |
| 2694 | chat - sorry - weeks - technology - phone | 36 | 2694_chat_sorry_weeks_technology |
| 2695 | friend - welcome - community - luck - new | 36 | 2695_friend_welcome_community_luck |
| 2696 | half - year - ones - order - questions | 36 | 2696_half_year_ones_order |
| 2697 | accounts - option - single - benefits - telegram | 36 | 2697_accounts_option_single_benefits |
| 2698 | holders - links - exchange - swap - supply | 36 | 2698_holders_links_exchange_swap |
| 2699 | great - day - question - amazing - today | 36 | 2699_great_day_question_amazing |
| 2700 | order - win - rest - case - latest | 36 | 2700_order_win_rest_case |
| 2701 | users - user - alot - lot - number | 36 | 2701_users_user_alot_lot |
| 2702 | tomorrow - better - yesterday - today - situation | 36 | 2702_tomorrow_better_yesterday_today |
| 2703 | market - - - - | 36 | 2703_market___ |
| 2704 | legit - successful - opportunity - happy - trade | 36 | 2704_legit_successful_opportunity_happy |
| 2705 | interested - daily - friends - platform - profits | 36 | 2705_interested_daily_friends_platform |
| 2706 | version - real - example - great - bitcoin | 36 | 2706_version_real_example_great |
| 2707 | tho - interesting - low - guy - possible | 36 | 2707_tho_interesting_low_guy |
| 2708 | dude - lots - buddy - ones - excited | 36 | 2708_dude_lots_buddy_ones |
| 2709 | help - idea - sorry - details - sure | 36 | 2709_help_idea_sorry_details |
| 2710 | second - type - job - perfect - worth | 36 | 2710_second_type_job_perfect |
| 2711 | crypto - digital - cryptocurrency - key - problems | 36 | 2711_crypto_digital_cryptocurrency_key |
| 2712 | news - market - chart - trade - new | 36 | 2712_news_market_chart_trade |
| 2713 | feature - normal - yesterday - shit - game | 36 | 2713_feature_normal_yesterday_shit |
| 2714 | rest - term - info - marketing - high | 36 | 2714_rest_term_info_marketing |
| 2715 | feature - major - upcoming - ecosystem - projects | 36 | 2715_feature_major_upcoming_ecosystem |
| 2716 | chat - public - group - price - rules | 36 | 2716_chat_public_group_price |
| 2717 | chance - life - experience - people - hit | 36 | 2717_chance_life_experience_people |
| 2718 | pool - cool - period - old - shit | 36 | 2718_pool_cool_period_old |
| 2719 | positive - life - work - people - good | 36 | 2719_positive_life_work_people |
| 2720 | mate - amazing - glad - positive - rate | 36 | 2720_mate_amazing_glad_positive |
| 2721 | situation - mind - buddy - shit - bad | 36 | 2721_situation_mind_buddy_shit |
| 2722 | dude - lots - little - positive - kind | 36 | 2722_dude_lots_little_positive |
| 2723 | legit - fee - withdrawal - withdraw - bitcoin | 36 | 2723_legit_fee_withdrawal_withdraw |
| 2724 | country - home - latest - binance - report | 36 | 2724_country_home_latest_binance |
| 2725 | official - chain - trust - answer - problem | 36 | 2725_official_chain_trust_answer |
| 2726 | release - head - article - control - example | 36 | 2726_release_head_article_control |
| 2727 | chance - answer - pool - ecosystem - fine | 36 | 2727_chance_answer_pool_ecosystem |
| 2728 | defi - number - supply - total - higher | 36 | 2728_defi_number_supply_total |
| 2729 | half - bitcoin - blockchain - special - quick | 36 | 2729_half_bitcoin_blockchain_special |
| 2730 | period - certain - worth - tomorrow - cool | 36 | 2730_period_certain_worth_tomorrow |
| 2731 | interested - home - daily - single - hour | 36 | 2731_interested_home_daily_single |
| 2732 | person - number - bitcoin - wallet - | 36 | 2732_person_number_bitcoin_wallet |
| 2733 | fun - hope - time - - | 36 | 2733_fun_hope_time_ |
| 2734 | sense - guy - option - cool - home | 36 | 2734_sense_guy_option_cool |
| 2735 | extra - pm - profit - daily - message | 36 | 2735_extra_pm_profit_daily |
| 2736 | bad - normal - point - luck - stuff | 36 | 2736_bad_normal_point_luck |
| 2737 | answer - type - couple - previous - solution | 36 | 2737_answer_type_couple_previous |
| 2738 | special - past - legit - experience - platform | 36 | 2738_special_past_legit_experience |
| 2739 | pepe - excited - ecosystem - exciting - latest | 36 | 2739_pepe_excited_ecosystem_exciting |
| 2740 | kind - step - comment - questions - recent | 36 | 2740_kind_step_comment_questions |
| 2741 | account - user - open - search - reason | 36 | 2741_account_user_open_search |
| 2742 | certain - report - fees - power - bitcoin | 36 | 2742_certain_report_fees_power |
| 2743 | important - hard - work - right - moment | 36 | 2743_important_hard_work_right |
| 2744 | transactions - lot - profit - payment - support | 36 | 2744_transactions_lot_profit_payment |
| 2745 | daily - profit - code - term - date | 36 | 2745_daily_profit_code_term |
| 2746 | sell - business - account - - | 36 | 2746_sell_business_account_ |
| 2747 | global - strong - opportunity - project - today | 36 | 2747_global_strong_opportunity_project |
| 2748 | similar - damn - certain - situation - difference | 36 | 2748_similar_damn_certain_situation |
| 2749 | interested - earn - hours - ways - simple | 36 | 2749_interested_earn_hours_ways |
| 2750 | place - good - - - | 36 | 2750_place_good__ |
| 2751 | total - buy - week - price - chart | 36 | 2751_total_buy_week_price |
| 2752 | assets - real - tbh - ecosystem - world | 36 | 2752_assets_real_tbh_ecosystem |
| 2753 | games - months - issues - major - way | 36 | 2753_games_months_issues_major |
| 2754 | cause - phone - multiple - normal - mate | 36 | 2754_cause_phone_multiple_normal |
| 2755 | turn - cause - glad - stuff - shit | 36 | 2755_turn_cause_glad_stuff |
| 2756 | awesome - channel - sure - people - money | 36 | 2756_awesome_channel_sure_people |
| 2757 | person - sorry - things - fact - example | 36 | 2757_person_sorry_things_fact |
| 2758 | connect - claim - open - fees - free | 36 | 2758_connect_claim_open_fees |
| 2759 | long - past - guy - bit - second | 36 | 2759_long_past_guy_bit |
| 2760 | complete - earlier - specific - yesterday - minutes | 36 | 2760_complete_earlier_specific_yesterday |
| 2761 | word - benefits - ai - little - legit | 36 | 2761_word_benefits_ai_little |
| 2762 | form - number - similar - question - likely | 36 | 2762_form_number_similar_question |
| 2763 | wonder - lol - network - sure - thanks | 36 | 2763_wonder_lol_network_sure |
| 2764 | crypto - type - update - biggest - clear | 36 | 2764_crypto_type_update_biggest |
| 2765 | news - btc - ethereum - bitcoin - set | 36 | 2765_news_btc_ethereum_bitcoin |
| 2766 | total - new - month - - | 36 | 2766_total_new_month_ |
| 2767 | email - support - links - option - chat | 36 | 2767_email_support_links_option |
| 2768 | simple - term - short - legit - plans | 36 | 2768_simple_term_short_legit |
| 2769 | man - info - - - | 36 | 2769_man_info__ |
| 2770 | words - example - withdrawal - successful - plans | 36 | 2770_words_example_withdrawal_successful |
| 2771 | god - couple - late - cause - face | 36 | 2771_god_couple_late_cause |
| 2772 | strong - level - possible - long - chart | 36 | 2772_strong_level_possible_long |
| 2773 | feature - ethereum - app - protocol - new | 36 | 2773_feature_ethereum_app_protocol |
| 2774 | easy - week - great - things - trading | 35 | 2774_easy_week_great_things |
| 2775 | ethereum - level - price - week - bitcoin | 35 | 2775_ethereum_level_price_week |
| 2776 | sure - bitcoin - future - country - recent | 35 | 2776_sure_bitcoin_future_country |
| 2777 | tho - sense - related - mate - glad | 35 | 2777_tho_sense_related_mate |
| 2778 | great - nice - good - day - week | 35 | 2778_great_nice_good_day |
| 2779 | - - - - | 35 | 2779____ |
| 2780 | nice - related - old - crazy - bad | 35 | 2780_nice_related_old_crazy |
| 2781 | protocol - smart - contract - possible - dev | 35 | 2781_protocol_smart_contract_possible |
| 2782 | wrong - damn - ppl - issues - dude | 35 | 2782_wrong_damn_ppl_issues |
| 2783 | things - term - months - news - worth | 35 | 2783_things_term_months_news |
| 2784 | contract - strong - potential - buy - binance | 35 | 2784_contract_strong_potential_buy |
| 2785 | participate - simple - half - early - sense | 35 | 2785_participate_simple_half_early |
| 2786 | sense - crazy - huge - supply - ways | 35 | 2786_sense_crazy_huge_supply |
| 2787 | likely - market - chart - funds - coin | 35 | 2787_likely_market_chart_funds |
| 2788 | certain - fun - pretty - games - stuff | 35 | 2788_certain_fun_pretty_games |
| 2789 | technology - biggest - glad - network - head | 35 | 2789_technology_biggest_glad_network |
| 2790 | nice - cool - building - interested - date | 35 | 2790_nice_cool_building_interested |
| 2791 | page - message - great - today - free | 35 | 2791_page_message_great_today |
| 2792 | wrong - dude - crazy - things - guys | 35 | 2792_wrong_dude_crazy_things |
| 2793 | certain - thinking - man - sir - | 35 | 2793_certain_thinking_man_sir |
| 2794 | members - group - real - main - fake | 35 | 2794_members_group_real_main |
| 2795 | digital - blockchain - global - network - protocol | 35 | 2795_digital_blockchain_global_network |
| 2796 | guys - profit - big - market - best | 35 | 2796_guys_profit_big_market |
| 2797 | tho - alot - ones - position - unique | 35 | 2797_tho_alot_ones_position |
| 2798 | thanks - crazy - bit - ama - old | 35 | 2798_thanks_crazy_bit_ama |
| 2799 | chance - assets - official - new - world | 35 | 2799_chance_assets_official_new |
| 2800 | near - project - strong - interesting - opinion | 35 | 2800_near_project_strong_interesting |
| 2801 | shit - biggest - tho - lol - rest | 35 | 2801_shit_biggest_tho_lol |
| 2802 | action - message - person - fake - | 35 | 2802_action_message_person_fake |
| 2803 | - - - - | 35 | 2803____ |
| 2804 | btc - point - shit - guy - ai | 35 | 2804_btc_point_shit_guy |
| 2805 | security - article - major - news - global | 35 | 2805_security_article_major_news |
| 2806 | upcoming - higher - message - accounts - app | 35 | 2806_upcoming_higher_message_accounts |
| 2807 | issues - trade - assets - link - bitcoin | 35 | 2807_issues_trade_assets_link |
| 2808 | single - difference - simple - list - specific | 35 | 2808_single_difference_simple_list |
| 2809 | dude - clear - correct - case - term | 35 | 2809_dude_clear_correct_case |
| 2810 | info - month - dm - team - | 35 | 2810_info_month_dm_team |
| 2811 | previous - past - near - head - turn | 35 | 2811_previous_past_near_head |
| 2812 | group - money - new - friend - glad | 35 | 2812_group_money_new_friend |
| 2813 | months - crazy - nft - crypto - certain | 35 | 2813_months_crazy_nft_crypto |
| 2814 | binance - live - features - similar - games | 35 | 2814_binance_live_features_similar |
| 2815 | similar - - - - | 35 | 2815_similar___ |
| 2816 | exchange - page - message - dm - | 35 | 2816_exchange_page_message_dm |
| 2817 | rate - admin - big - low - interested | 35 | 2817_rate_admin_big_low |
| 2818 | questions - welcome - mate - free - thanks | 35 | 2818_questions_welcome_mate_free |
| 2819 | defi - links - option - wallet - media | 35 | 2819_defi_links_option_wallet |
| 2820 | hand - user - feature - interesting - data | 35 | 2820_hand_user_feature_interesting |
| 2821 | access - - - - | 35 | 2821_access___ |
| 2822 | leave - reason - question - man - lot | 35 | 2822_leave_reason_question_man |
| 2823 | showing - strong - digital - market - trade | 35 | 2823_showing_strong_digital_market |
| 2824 | unique - video - ready - network - nft | 35 | 2824_unique_video_ready_network |
| 2825 | feature - product - spot - news - experience | 35 | 2825_feature_product_spot_news |
| 2826 | word - face - shit - guy - cool | 35 | 2826_word_face_shit_guy |
| 2827 | global - public - user - issue - liquidity | 35 | 2827_global_public_user_issue |
| 2828 | contact - second - profit - yesterday - legit | 35 | 2828_contact_second_profit_yesterday |
| 2829 | total - volume - transaction - change - contract | 35 | 2829_total_volume_transaction_change |
| 2830 | video - comment - attention - problems - problem | 35 | 2830_video_comment_attention_problems |
| 2831 | ways - liquidity - crypto - investors - opinion | 35 | 2831_ways_liquidity_crypto_investors |
| 2832 | list - ways - lot - - | 35 | 2832_list_ways_lot_ |
| 2833 | minimum - deposit - daily - sign - term | 35 | 2833_minimum_deposit_daily_sign |
| 2834 | break - wish - excited - night - early | 35 | 2834_break_wish_excited_night |
| 2835 | night - links - media - end - users | 35 | 2835_night_links_media_end |
| 2836 | coin - coins - transactions - cryptocurrency - secure | 35 | 2836_coin_coins_transactions_cryptocurrency |
| 2837 | guide - key - wallet - wrong - account | 35 | 2837_guide_key_wallet_wrong |
| 2838 | starting - digital - trading - life - new | 35 | 2838_starting_digital_trading_life |
| 2839 | - - - - | 35 | 2839____ |
| 2840 | couple - public - special - man - today | 35 | 2840_couple_public_special_man |
| 2841 | post - rate - development - channel - head | 35 | 2841_post_rate_development_channel |
| 2842 | minutes - second - profit - pm - minimum | 35 | 2842_minutes_second_profit_pm |
| 2843 | bot - ai - trading - features - special | 35 | 2843_bot_ai_trading_features |
| 2844 | hand - near - difference - guy - man | 35 | 2844_hand_near_difference_guy |
| 2845 | fun - message - online - morning - want | 35 | 2845_fun_message_online_morning |
| 2846 | hours - phone - drop - pay - real | 35 | 2846_hours_phone_drop_pay |
| 2847 | plan - ready - - - | 35 | 2847_plan_ready__ |
| 2848 | problems - cause - personal - kind - opinion | 34 | 2848_problems_cause_personal_kind |
| 2849 | hard - work - money - order - smart | 34 | 2849_hard_work_money_order |
| 2850 | near - run - moment - excited - type | 34 | 2850_near_run_moment_excited |
| 2851 | night - head - rest - tomorrow - early | 34 | 2851_night_head_rest_tomorrow |
| 2852 | bitcoin - global - power - future - launch | 34 | 2852_bitcoin_global_power_future |
| 2853 | note - global - latest - power - digital | 34 | 2853_note_global_latest_power |
| 2854 | deal - welcome - contact - ready - interested | 34 | 2854_deal_welcome_contact_ready |
| 2855 | btc - level - game - market - near | 34 | 2855_btc_level_game_market |
| 2856 | position - buy - price - web - ecosystem | 34 | 2856_position_buy_price_web |
| 2857 | country - friend - great - - | 34 | 2857_country_friend_great_ |
| 2858 | word - moving - comment - example - recent | 34 | 2858_word_moving_comment_example |
| 2859 | sure - way - matter - looks - date | 34 | 2859_sure_way_matter_looks |
| 2860 | attention - private - swap - opportunity - bitcoin | 34 | 2860_attention_private_swap_opportunity |
| 2861 | space - guy - excited - possible - games | 34 | 2861_space_guy_excited_possible |
| 2862 | pls - option - way - correct - moment | 34 | 2862_pls_option_way_correct |
| 2863 | latest - report - possible - risk - clear | 34 | 2863_latest_report_possible_risk |
| 2864 | close - release - excited - attention - research | 34 | 2864_close_release_excited_attention |
| 2865 | position - chart - holders - funds - price | 34 | 2865_position_chart_holders_funds |
| 2866 | way - ways - case - run - lol | 34 | 2866_way_ways_case_run |
| 2867 | ama - utc - rules - twitter - pm | 34 | 2867_ama_utc_rules_twitter |
| 2868 | times - country - chance - small - little | 34 | 2868_times_country_chance_small |
| 2869 | post - life - code - - | 34 | 2869_post_life_code_ |
| 2870 | marketing - service - crypto - alot - positive | 34 | 2870_marketing_service_crypto_alot |
| 2871 | situation - order - example - night - morning | 34 | 2871_situation_order_example_night |
| 2872 | ones - kind - point - perfect - guys | 34 | 2872_ones_kind_point_perfect |
| 2873 | today - mate - guys - nice - team | 34 | 2873_today_mate_guys_nice |
| 2874 | guide - earlier - period - likely - details | 34 | 2874_guide_earlier_period_likely |
| 2875 | ethereum - bitcoin - - - | 34 | 2875_ethereum_bitcoin__ |
| 2876 | - - - - | 34 | 2876____ |
| 2877 | luck - dear - details - amazing - week | 34 | 2877_luck_dear_details_amazing |
| 2878 | today - - - - | 34 | 2878_today___ |
| 2879 | deal - money - ready - situation - rate | 34 | 2879_deal_money_ready_situation |
| 2880 | worth - tokens - rest - new - public | 34 | 2880_worth_tokens_rest_new |
| 2881 | home - extra - thanks - hours - help | 34 | 2881_home_extra_thanks_hours |
| 2882 | hit - massive - price - minutes - biggest | 34 | 2882_hit_massive_price_minutes |
| 2883 | major - binance - pay - services - exchange | 34 | 2883_major_binance_pay_services |
| 2884 | time - point - crazy - issues - kind | 34 | 2884_time_point_crazy_issues |
| 2885 | ai - news - month - space - | 34 | 2885_ai_news_month_space |
| 2886 | transaction - digital - trade - - | 34 | 2886_transaction_digital_trade_ |
| 2887 | project - second - - - | 34 | 2887_project_second__ |
| 2888 | home - opinion - clear - shows - try | 34 | 2888_home_opinion_clear_shows |
| 2889 | user - users - available - playing - night | 34 | 2889_user_users_available_playing |
| 2890 | dude - shit - luck - dev - guy | 34 | 2890_dude_shit_luck_dev |
| 2891 | secure - fee - opportunity - register - payment | 34 | 2891_secure_fee_opportunity_register |
| 2892 | security - online - person - data - end | 34 | 2892_security_online_person_data |
| 2893 | projects - lol - moment - project - | 34 | 2893_projects_lol_moment_project |
| 2894 | shit - run - tho - old - alot | 34 | 2894_shit_run_tho_old |
| 2895 | shit - single - country - morning - rate | 34 | 2895_shit_single_country_morning |
| 2896 | usdt - minimum - members - extra - friends | 34 | 2896_usdt_minimum_members_extra |
| 2897 | successful - huge - risk - things - thing | 34 | 2897_successful_huge_risk_things |
| 2898 | participate - event - reward - current - rewards | 34 | 2898_participate_event_reward_current |
| 2899 | tokens - token - guy - value - near | 34 | 2899_tokens_token_guy_value |
| 2900 | power - contract - world - community - massive | 34 | 2900_power_contract_world_community |
| 2901 | supply - total - current - friend - correct | 34 | 2901_supply_total_current_friend |
| 2902 | official - block - report - real - fake | 34 | 2902_official_block_report_real |
| 2903 | game - experience - web - ecosystem - cool | 34 | 2903_game_experience_web_ecosystem |
| 2904 | simple - bot - friends - share - withdrawal | 34 | 2904_simple_bot_friends_share |
| 2905 | bot - app - official - perfect - | 34 | 2905_bot_app_official_perfect |
| 2906 | damn - tbh - normal - solution - true | 34 | 2906_damn_tbh_normal_solution |
| 2907 | info - free - - - | 34 | 2907_info_free__ |
| 2908 | links - link - bad - country - job | 34 | 2908_links_link_bad_country |
| 2909 | number - week - network - days - bitcoin | 34 | 2909_number_week_network_days |
| 2910 | great - good - night - tomorrow - | 34 | 2910_great_good_night_tomorrow |
| 2911 | success - opinion - positive - way - questions | 34 | 2911_success_opinion_positive_way |
| 2912 | spot - - - - | 34 | 2912_spot___ |
| 2913 | unique - family - place - claim - ethereum | 34 | 2913_unique_family_place_claim |
| 2914 | ones - past - guy - man - similar | 34 | 2914_ones_past_guy_man |
| 2915 | excited - exciting - interested - launch - pool | 34 | 2915_excited_exciting_interested_launch |
| 2916 | similar - - - - | 34 | 2916_similar___ |
| 2917 | secure - transactions - live - protocol - ready | 34 | 2917_secure_transactions_live_protocol |
| 2918 | earlier - reply - attention - price - process | 34 | 2918_earlier_reply_attention_price |
| 2919 | form - point - course - interesting - early | 34 | 2919_form_point_course_interesting |
| 2920 | total - couple - current - hand - question | 34 | 2920_total_couple_current_hand |
| 2921 | guy - close - wonder - legit - late | 34 | 2921_guy_close_wonder_legit |
| 2922 | dev - ethereum - years - transactions - trade | 34 | 2922_dev_ethereum_years_transactions |
| 2923 | case - positive - yesterday - latest - situation | 34 | 2923_case_positive_yesterday_latest |
| 2924 | global - digital - half - article - past | 34 | 2924_global_digital_half_article |
| 2925 | multiple - report - risk - - | 34 | 2925_multiple_report_risk_ |
| 2926 | bot - type - available - channel - link | 34 | 2926_bot_type_available_channel |
| 2927 | unique - protocol - admin - fake - post | 34 | 2927_unique_protocol_admin_fake |
| 2928 | fake - cryptocurrency - ecosystem - power - year | 34 | 2928_fake_cryptocurrency_ecosystem_power |
| 2929 | online - minimum - sign - days - deposit | 34 | 2929_online_minimum_sign_days |
| 2930 | project - questions - - - | 34 | 2930_project_questions__ |
| 2931 | swap - announcement - complete - token - assets | 34 | 2931_swap_announcement_complete_token |
| 2932 | deal - game - token - - | 34 | 2932_deal_game_token_ |
| 2933 | dude - product - shit - guy - night | 34 | 2933_dude_product_shit_guy |
| 2934 | account - hour - clear - problem - different | 34 | 2934_account_hour_clear_problem |
| 2935 | mate - true - course - friend - words | 34 | 2935_mate_true_course_friend |
| 2936 | game - guide - access - use - new | 34 | 2936_game_guide_access_use |
| 2937 | link - happy - - - | 34 | 2937_link_happy__ |
| 2938 | life - people - amazing - thing - happy | 34 | 2938_life_people_amazing_thing |
| 2939 | article - product - correct - web - new | 34 | 2939_article_product_correct_web |
| 2940 | exciting - ai - event - transactions - ready | 34 | 2940_exciting_ai_event_transactions |
| 2941 | low - crazy - key - second - point | 34 | 2941_low_crazy_key_second |
| 2942 | head - home - guy - word - attention | 34 | 2942_head_home_guy_word |
| 2943 | attention - ecosystem - investors - token - tokens | 34 | 2943_attention_ecosystem_investors_token |
| 2944 | liquidity - solution - special - interesting - upcoming | 34 | 2944_liquidity_solution_special_interesting |
| 2945 | network - blockchain - technology - transactions - secure | 34 | 2945_network_blockchain_technology_transactions |
| 2946 | minimum - hours - deposit - withdrawal - new | 34 | 2946_minimum_hours_deposit_withdrawal |
| 2947 | break - job - old - end - place | 34 | 2947_break_job_old_end |
| 2948 | plans - business - funds - launch - new | 34 | 2948_plans_business_funds_launch |
| 2949 | ama - post - main - power - long | 33 | 2949_ama_post_main_power |
| 2950 | online - mate - lots - available - days | 33 | 2950_online_mate_lots_available |
| 2951 | end - risk - level - official - daily | 33 | 2951_end_risk_level_official |
| 2952 | similar - phone - stuff - wish - issues | 33 | 2952_similar_phone_stuff_wish |
| 2953 | technology - early - cryptocurrency - crypto - liquidity | 33 | 2953_technology_early_cryptocurrency_crypto |
| 2954 | problems - ai - blockchain - real - future | 33 | 2954_problems_ai_blockchain_real |
| 2955 | special - search - friend - chance - man | 33 | 2955_special_search_friend_chance |
| 2956 | profits - huge - lots - legit - friends | 33 | 2956_profits_huge_lots_legit |
| 2957 | words - word - minutes - game - date | 33 | 2957_words_word_minutes_game |
| 2958 | security - accounts - assets - digital - user | 33 | 2958_security_accounts_assets_digital |
| 2959 | block - specific - security - question - user | 33 | 2959_block_specific_security_question |
| 2960 | number - months - bitcoin - news - | 33 | 2960_number_months_bitcoin_news |
| 2961 | transaction - binance - fees - week - btc | 33 | 2961_transaction_binance_fees_week |
| 2962 | main - official - chat - admin - welcome | 33 | 2962_main_official_chat_admin |
| 2963 | thinking - plan - code - chance - question | 33 | 2963_thinking_plan_code_chance |
| 2964 | face - sir - amazing - lol - way | 33 | 2964_face_sir_amazing_lol |
| 2965 | sir - wait - yesterday - upcoming - hours | 33 | 2965_sir_wait_yesterday_upcoming |
| 2966 | contact - team - email - interested - thanks | 33 | 2966_contact_team_email_interested |
| 2967 | money - early - likely - example - crazy | 33 | 2967_money_early_likely_example |
| 2968 | higher - perfect - btc - price - bitcoin | 33 | 2968_higher_perfect_btc_price |
| 2969 | fast - available - exchange - - | 33 | 2969_fast_available_exchange_ |
| 2970 | security - smart - experience - chain - transaction | 33 | 2970_security_smart_experience_chain |
| 2971 | business - profit - week - investors - days | 33 | 2971_business_profit_week_investors |
| 2972 | sure - days - - - | 33 | 2972_sure_days__ |
| 2973 | comment - launch - fair - pepe - hour | 33 | 2973_comment_launch_fair_pepe |
| 2974 | opinion - public - app - single - issues | 33 | 2974_opinion_public_app_single |
| 2975 | similar - wrong - night - country - nice | 33 | 2975_similar_wrong_night_country |
| 2976 | glad - little - happy - pay - month | 33 | 2976_glad_little_happy_pay |
| 2977 | btw - friend - action - place - | 33 | 2977_btw_friend_action_place |
| 2978 | opportunity - just - investment - hours - home | 33 | 2978_opportunity_just_investment_hours |
| 2979 | idea - sure - mind - information - | 33 | 2979_idea_sure_mind_information |
| 2980 | protocol - benefits - opportunity - hours - | 33 | 2980_protocol_benefits_opportunity_hours |
| 2981 | morning - good - - - | 33 | 2981_morning_good__ |
| 2982 | trade - time - weeks - good - couple | 33 | 2982_trade_time_weeks_good |
| 2983 | cool - exciting - current - amazing - update | 33 | 2983_cool_exciting_current_amazing |
| 2984 | things - sure - team - - | 33 | 2984_things_sure_team_ |
| 2985 | feature - form - click - details - head | 33 | 2985_feature_form_click_details |
| 2986 | major - higher - active - bitcoin - hope | 33 | 2986_major_higher_active_bitcoin |
| 2987 | normal - bit - difference - fact - close | 33 | 2987_normal_bit_difference_fact |
| 2988 | face - deal - type - ethereum - country | 33 | 2988_face_deal_type_ethereum |
| 2989 | short - - - - | 33 | 2989_short___ |
| 2990 | action - success - things - life - way | 33 | 2990_action_success_things_life |
| 2991 | risk - level - official - daily - | 33 | 2991_risk_level_official_daily |
| 2992 | related - video - global - financial - product | 33 | 2992_related_video_global_financial |
| 2993 | previous - holders - private - launch - huge | 33 | 2993_previous_holders_private_launch |
| 2994 | holders - opinion - holding - coins - past | 33 | 2994_holders_opinion_holding_coins |
| 2995 | defi - assets - process - risk - trading | 33 | 2995_defi_assets_process_risk |
| 2996 | certain - control - security - users - risk | 33 | 2996_certain_control_security_users |
| 2997 | similar - dude - luck - technology - nice | 33 | 2997_similar_dude_luck_technology |
| 2998 | similar - fact - past - ones - word | 33 | 2998_similar_fact_past_ones |
| 2999 | plan - minimum - super - site - successful | 33 | 2999_plan_minimum_super_site |
| 3000 | money - way - work - stuff - head | 33 | 3000_money_way_work_stuff |
| 3001 | blockchain - positive - benefits - global - assets | 33 | 3001_blockchain_positive_benefits_global |
| 3002 | simple - wrong - man - welcome - real | 33 | 3002_simple_wrong_man_welcome |
| 3003 | sense - dude - product - interesting - night | 33 | 3003_sense_dude_product_interesting |
| 3004 | personal - possible - increase - crypto - worth | 33 | 3004_personal_possible_increase_crypto |
| 3005 | ai - unique - interesting - different - happy | 33 | 3005_ai_unique_interesting_different |
| 3006 | certain - head - clear - news - guy | 33 | 3006_certain_head_clear_news |
| 3007 | open - token - - - | 33 | 3007_open_token__ |
| 3008 | smart - contract - safe - - | 33 | 3008_smart_contract_safe_ |
| 3009 | guy - dear - rest - lol - person | 33 | 3009_guy_dear_rest_lol |
| 3010 | projects - project - building - product - fact | 33 | 3010_projects_project_building_product |
| 3011 | pls - value - price - case - rate | 33 | 3011_pls_value_price_case |
| 3012 | holders - cryptocurrency - launch - bitcoin - single | 33 | 3012_holders_cryptocurrency_launch_bitcoin |
| 3013 | real - fun - single - normal - dm | 33 | 3013_real_fun_single_normal |
| 3014 | - - - - | 33 | 3014____ |
| 3015 | cryptocurrency - value - money - certain - latest | 33 | 3015_cryptocurrency_value_money_certain |
| 3016 | technology - ways - blockchain - development - important | 33 | 3016_technology_ways_blockchain_development |
| 3017 | legit - spot - difference - information - trading | 33 | 3017_legit_spot_difference_information |
| 3018 | fine - simple - single - features - guide | 33 | 3018_fine_simple_single_features |
| 3019 | fair - situation - playing - brother - special | 33 | 3019_fair_situation_playing_brother |
| 3020 | example - number - term - short - issue | 33 | 3020_example_number_term_short |
| 3021 | total - value - change - transactions - update | 33 | 3021_total_value_change_transactions |
| 3022 | opinion - app - issues - solution - public | 33 | 3022_opinion_app_issues_solution |
| 3023 | normal - wallet - couple - version - contract | 33 | 3023_normal_wallet_couple_version |
| 3024 | ready - guys - home - time - | 33 | 3024_ready_guys_home_time |
| 3025 | option - higher - games - longer - game | 33 | 3025_option_higher_games_longer |
| 3026 | plans - plan - team - months - major | 33 | 3026_plans_plan_team_months |
| 3027 | personal - admin - bot - telegram - support | 33 | 3027_personal_admin_bot_telegram |
| 3028 | specific - recent - - - | 33 | 3028_specific_recent__ |
| 3029 | fix - hour - normal - update - luck | 33 | 3029_fix_hour_normal_update |
| 3030 | point - tho - correct - positive - hour | 33 | 3030_point_tho_correct_positive |
| 3031 | words - guys - times - amazing - online | 33 | 3031_words_guys_times_amazing |
| 3032 | day - good - - - | 33 | 3032_day_good__ |
| 3033 | interested - single - years - - | 33 | 3033_interested_single_years_ |
| 3034 | person - life - guys - important - amazing | 33 | 3034_person_life_guys_important |
| 3035 | certain - previous - assets - event - worth | 33 | 3035_certain_previous_assets_event |
| 3036 | protocol - deal - ethereum - transactions - twitter | 33 | 3036_protocol_deal_ethereum_transactions |
| 3037 | feature - phone - control - tho - little | 32 | 3037_feature_phone_control_tho |
| 3038 | low - high - listing - btc - friend | 32 | 3038_low_high_listing_btc |
| 3039 | project - profit - attention - risk - investors | 32 | 3039_project_profit_attention_risk |
| 3040 | phone - bit - normal - difference - cause | 32 | 3040_phone_bit_normal_difference |
| 3041 | problems - feature - clear - reason - option | 32 | 3041_problems_feature_clear_reason |
| 3042 | benefits - interested - friends - fun - real | 32 | 3042_benefits_interested_friends_fun |
| 3043 | comment - twitter - official - post - page | 32 | 3043_comment_twitter_official_post |
| 3044 | network - wallet - - - | 32 | 3044_network_wallet__ |
| 3045 | channel - address - message - tokens - token | 32 | 3045_channel_address_message_tokens |
| 3046 | updates - update - ppl - team - nice | 32 | 3046_updates_update_ppl_team |
| 3047 | certain - global - normal - glad - past | 32 | 3047_certain_global_normal_glad |
| 3048 | fees - major - transaction - year - btc | 32 | 3048_fees_major_transaction_year |
| 3049 | situation - message - user - join - admin | 32 | 3049_situation_message_user_join |
| 3050 | successful - big - group - - | 32 | 3050_successful_big_group_ |
| 3051 | term - long - account - team - | 32 | 3051_term_long_account_team |
| 3052 | cryptocurrency - financial - opinion - example - value | 32 | 3052_cryptocurrency_financial_opinion_example |
| 3053 | dude - tho - awesome - rest - idea | 32 | 3053_dude_tho_awesome_rest |
| 3054 | weeks - sense - major - reason - dev | 32 | 3054_weeks_sense_major_reason |
| 3055 | contact - form - email - course - thanks | 32 | 3055_contact_form_email_course |
| 3056 | guy - similar - bad - end - course | 32 | 3056_guy_similar_bad_end |
| 3057 | multiple - games - personal - rules - fun | 32 | 3057_multiple_games_personal_rules |
| 3058 | reply - single - help - user - change | 32 | 3058_reply_single_help_user |
| 3059 | user - price - worth - case - company | 32 | 3059_user_price_worth_case |
| 3060 | guy - earlier - country - dude - sorry | 32 | 3060_guy_earlier_country_dude |
| 3061 | technology - block - change - year - transaction | 32 | 3061_technology_block_change_year |
| 3062 | job - crypto - year - stuff - word | 32 | 3062_job_crypto_year_stuff |
| 3063 | difference - home - success - platform - love | 32 | 3063_difference_home_success_platform |
| 3064 | half - likely - answer - business - opinion | 32 | 3064_half_likely_answer_business |
| 3065 | holders - volume - chart - tg - block | 32 | 3065_holders_volume_chart_tg |
| 3066 | current - chain - supply - technology - swap | 32 | 3066_current_chain_supply_technology |
| 3067 | launch - global - fees - marketing - contract | 32 | 3067_launch_global_fees_marketing |
| 3068 | risk - btc - fees - price - bitcoin | 32 | 3068_risk_btc_fees_price |
| 3069 | situation - important - life - thing - person | 32 | 3069_situation_important_life_thing |
| 3070 | simple - space - lot - money - | 32 | 3070_simple_space_lot_money |
| 3071 | reason - day - way - big - help | 32 | 3071_reason_day_way_big |
| 3072 | interesting - thing - sense - stuff - shit | 32 | 3072_interesting_thing_sense_stuff |
| 3073 | tokens - normal - token - address - problems | 32 | 3073_tokens_normal_token_address |
| 3074 | day - good - dear - nice - great | 32 | 3074_day_good_dear_nice |
| 3075 | normal - break - solution - buddy - close | 32 | 3075_normal_break_solution_buddy |
| 3076 | coins - coin - example - short - trade | 32 | 3076_coins_coin_example_short |
| 3077 | new - - - - | 32 | 3077_new___ |
| 3078 | security - step - wallet - follow - personal | 32 | 3078_security_step_wallet_follow |
| 3079 | trading - trade - investment - - | 32 | 3079_trading_trade_investment_ |
| 3080 | contract - address - token - correct - details | 32 | 3080_contract_address_token_correct |
| 3081 | business - special - dev - potential - chain | 32 | 3081_business_special_dev_potential |
| 3082 | product - case - months - trading - years | 32 | 3082_product_case_months_trading |
| 3083 | kind - safe - benefits - members - technology | 32 | 3083_kind_safe_benefits_members |
| 3084 | sense - damn - couple - luck - crazy | 32 | 3084_sense_damn_couple_luck |
| 3085 | deal - worth - days - true - | 32 | 3085_deal_worth_days_true |
| 3086 | media - social - - - | 32 | 3086_media_social__ |
| 3087 | technology - building - world - bitcoin - development | 32 | 3087_technology_building_world_bitcoin |
| 3088 | words - home - ethereum - trading - info | 32 | 3088_words_home_ethereum_trading |
| 3089 | updates - plans - new - global - day | 32 | 3089_updates_plans_new_global |
| 3090 | example - binance - email - claim - daily | 32 | 3090_example_binance_email_claim |
| 3091 | chat - real - date - dm - normal | 32 | 3091_chat_real_date_dm |
| 3092 | glad - kind - nice - thing - friends | 32 | 3092_glad_kind_nice_thing |
| 3093 | type - buddy - correct - unique - friends | 32 | 3093_type_buddy_correct_unique |
| 3094 | coin - questions - ways - interesting - important | 32 | 3094_coin_questions_ways_interesting |
| 3095 | latest - official - dev - shit - site | 32 | 3095_latest_official_dev_shit |
| 3096 | wallet - mate - web - version - features | 32 | 3096_wallet_mate_web_version |
| 3097 | ones - important - contract - address - sure | 32 | 3097_ones_important_contract_address |
| 3098 | digital - share - year - projects - group | 32 | 3098_digital_share_year_projects |
| 3099 | launch - utc - great - today - guys | 32 | 3099_launch_utc_great_today |
| 3100 | link - investment - - - | 32 | 3100_link_investment__ |
| 3101 | yesterday - tomorrow - bad - man - little | 32 | 3101_yesterday_tomorrow_bad_man |
| 3102 | hour - solution - business - accounts - deal | 32 | 3102_hour_solution_business_accounts |
| 3103 | higher - comment - luck - night - spot | 32 | 3103_higher_comment_luck_night |
| 3104 | problems - contact - step - successful - withdrawal | 32 | 3104_problems_contact_step_successful |
| 3105 | main - example - solution - games - version | 32 | 3105_main_example_solution_games |
| 3106 | cause - report - risk - - | 32 | 3106_cause_report_risk_ |
| 3107 | fair - option - control - ethereum - kind | 32 | 3107_fair_option_control_ethereum |
| 3108 | super - shit - head - opinion - biggest | 32 | 3108_super_shit_head_opinion |
| 3109 | issue - deal - announcement - big - sir | 32 | 3109_issue_deal_announcement_big |
| 3110 | supply - details - pool - total - update | 32 | 3110_supply_details_pool_total |
| 3111 | job - yesterday - shit - cool - home | 32 | 3111_job_yesterday_shit_cool |
| 3112 | man - stuff - ones - home - wan | 32 | 3112_man_stuff_ones_home |
| 3113 | website - available - - - | 32 | 3113_website_available__ |
| 3114 | data - second - digital - trade - bitcoin | 32 | 3114_data_second_digital_trade |
| 3115 | volume - holders - supply - market - price | 32 | 3115_volume_holders_supply_market |
| 3116 | face - head - point - late - example | 32 | 3116_face_head_point_late |
| 3117 | ones - drop - profit - profits - crypto | 32 | 3117_ones_drop_profit_profits |
| 3118 | supply - total - correct - admin - current | 32 | 3118_supply_total_correct_admin |
| 3119 | market - normal - past - bit - morning | 32 | 3119_market_normal_past_bit |
| 3120 | special - week - nft - night - fun | 32 | 3120_special_week_nft_night |
| 3121 | product - defi - earn - wallet - link | 32 | 3121_product_defi_earn_wallet |
| 3122 | head - sense - face - buddy - fair | 32 | 3122_head_sense_face_buddy |
| 3123 | ready - market - crypto - control - difference | 32 | 3123_ready_market_crypto_control |
| 3124 | glad - little - guy - sorry - friends | 32 | 3124_glad_little_guy_sorry |
| 3125 | glad - news - great - time - | 32 | 3125_glad_news_great_time |
| 3126 | safe - - - - | 32 | 3126_safe___ |
| 3127 | buy - issue - chain - tokens - profits | 32 | 3127_buy_issue_chain_tokens |
| 3128 | minutes - case - fine - chain - hour | 32 | 3128_minutes_case_fine_chain |
| 3129 | opinion - bro - clear - online - issues | 32 | 3129_opinion_bro_clear_online |
| 3130 | glad - latest - bro - channel - great | 32 | 3130_glad_latest_bro_channel |
| 3131 | morning - mate - happy - good - nice | 32 | 3131_morning_mate_happy_good |
| 3132 | damn - ppl - dude - option - fine | 32 | 3132_damn_ppl_dude_option |
| 3133 | guy - brother - situation - live - thing | 32 | 3133_guy_brother_situation_live |
| 3134 | digital - assets - space - ways - volume | 32 | 3134_digital_assets_space_ways |
| 3135 | business - message - - - | 32 | 3135_business_message__ |
| 3136 | fair - sense - head - weeks - case | 32 | 3136_fair_sense_head_weeks |
| 3137 | protocol - discord - dev - news - updates | 32 | 3137_protocol_discord_dev_news |
| 3138 | fee - specific - fees - withdrawal - withdraw | 32 | 3138_fee_specific_fees_withdrawal |
| 3139 | market - attention - lot - normal - control | 32 | 3139_market_attention_lot_normal |
| 3140 | playing - second - end - time - | 32 | 3140_playing_second_end_time |
| 3141 | near - time - night - second - space | 32 | 3141_near_time_night_second |
| 3142 | fake - attention - block - admin - accounts | 32 | 3142_fake_attention_block_admin |
| 3143 | likely - tho - cool - bro - life | 32 | 3143_likely_tho_cool_bro |
| 3144 | late - close - control - fact - related | 32 | 3144_late_close_control_fact |
| 3145 | multiple - launch - private - transaction - listing | 32 | 3145_multiple_launch_private_transaction |
| 3146 | fee - fees - certain - opinion - chain | 32 | 3146_fee_fees_certain_opinion |
| 3147 | moment - day - sounds - mate - buddy | 32 | 3147_moment_day_sounds_mate |
| 3148 | pm - new - - - | 32 | 3148_pm_new__ |
| 3149 | exchange - access - users - crypto - major | 32 | 3149_exchange_access_users_crypto |
| 3150 | ama - bot - user - dear - digital | 32 | 3150_ama_bot_user_dear |
| 3151 | visit - contract - rules - protocol - latest | 32 | 3151_visit_contract_rules_protocol |
| 3152 | chat - video - bot - group - question | 32 | 3152_chat_video_bot_group |
| 3153 | late - website - ama - upcoming - total | 32 | 3153_late_website_ama_upcoming |
| 3154 | order - course - questions - important - worth | 32 | 3154_order_course_questions_important |
| 3155 | guy - brother - word - man - night | 31 | 3155_guy_brother_word_man |
| 3156 | btc - price - bitcoin - current - profit | 31 | 3156_btc_price_bitcoin_current |
| 3157 | comment - tomorrow - chance - launch - time | 31 | 3157_comment_tomorrow_chance_launch |
| 3158 | chance - cause - answer - second - person | 31 | 3158_chance_cause_answer_second |
| 3159 | shit - term - business - pool - true | 31 | 3159_shit_term_business_pool |
| 3160 | face - amazing - way - payment - best | 31 | 3160_face_amazing_way_payment |
| 3161 | official - block - links - fake - announcement | 31 | 3161_official_block_links_fake |
| 3162 | risk - level - official - daily - | 31 | 3162_risk_level_official_daily |
| 3163 | special - true - successful - trade - game | 31 | 3163_special_true_successful_trade |
| 3164 | problems - site - address - email - coins | 31 | 3164_problems_site_address_email |
| 3165 | utc - period - deposit - withdrawal - days | 31 | 3165_utc_period_deposit_withdrawal |
| 3166 | different - face - research - words - person | 31 | 3166_different_face_research_words |
| 3167 | hand - near - hope - past - end | 31 | 3167_hand_near_hope_past |
| 3168 | defi - capital - bitcoin - wallet - major | 31 | 3168_defi_capital_bitcoin_wallet |
| 3169 | position - job - development - open - | 31 | 3169_position_job_development_open |
| 3170 | link - version - investment - protocol - | 31 | 3170_link_version_investment_protocol |
| 3171 | country - na - friends - sure - people | 31 | 3171_country_na_friends_sure |
| 3172 | deal - sounds - alright - couple - bad | 31 | 3172_deal_sounds_alright_couple |
| 3173 | games - chart - holders - funds - website | 31 | 3173_games_chart_holders_funds |
| 3174 | fees - fee - trading - spot - position | 31 | 3174_fees_fee_trading_spot |
| 3175 | news - game - channel - community - video | 31 | 3175_news_game_channel_community |
| 3176 | upcoming - investment - - - | 31 | 3176_upcoming_investment__ |
| 3177 | dude - shit - home - bit - answer | 31 | 3177_dude_shit_home_bit |
| 3178 | rules - claim - transactions - worth - minimum | 31 | 3178_rules_claim_transactions_worth |
| 3179 | fair - rate - price - normal - contract | 31 | 3179_fair_rate_price_normal |
| 3180 | coin - list - coins - spot - interested | 31 | 3180_coin_list_coins_spot |
| 3181 | multiple - interesting - trading - spot - specific | 31 | 3181_multiple_interesting_trading_spot |
| 3182 | chat - interested - big - dm - | 31 | 3182_chat_interested_big_dm |
| 3183 | alot - sir - clear - interesting - night | 31 | 3183_alot_sir_clear_interesting |
| 3184 | hour - guide - binance - message - capital | 31 | 3184_hour_guide_binance_message |
| 3185 | strong - successful - times - win - happy | 31 | 3185_strong_successful_times_win |
| 3186 | brother - special - buddy - night - fine | 31 | 3186_brother_special_buddy_night |
| 3187 | kinda - assets - deal - real - legit | 31 | 3187_kinda_assets_deal_real |
| 3188 | clear - similar - late - thing - admin | 31 | 3188_clear_similar_late_thing |
| 3189 | half - home - game - simple - rules | 31 | 3189_half_home_game_simple |
| 3190 | times - risk - trust - success - service | 31 | 3190_times_risk_trust_success |
| 3191 | questions - huge - head - market - recent | 31 | 3191_questions_huge_head_market |
| 3192 | recent - amazing - digital - quick - doubt | 31 | 3192_recent_amazing_digital_quick |
| 3193 | event - amazing - community - today - day | 31 | 3193_event_amazing_community_today |
| 3194 | share - extra - exchange - coin - home | 31 | 3194_share_extra_exchange_coin |
| 3195 | head - security - fair - ai - use | 31 | 3195_head_security_fair_ai |
| 3196 | future - - - - | 31 | 3196_future___ |
| 3197 | attention - research - information - small - problem | 31 | 3197_attention_research_information_small |
| 3198 | experience - register - group - investment - company | 31 | 3198_experience_register_group_investment |
| 3199 | create - game - digital - building - home | 31 | 3199_create_game_digital_building |
| 3200 | cryptocurrency - market - high - bitcoin - | 31 | 3200_cryptocurrency_market_high_bitcoin |
| 3201 | block - private - likely - security - user | 31 | 3201_block_private_likely_security |
| 3202 | guy - old - fun - year - couple | 31 | 3202_guy_old_fun_year |
| 3203 | case - email - reply - update - support | 31 | 3203_case_email_reply_update |
| 3204 | page - message - dm - - | 31 | 3204_page_message_dm_ |
| 3205 | left - super - news - phone - smart | 31 | 3205_left_super_news_phone |
| 3206 | rewards - spot - details - trading - deposit | 31 | 3206_rewards_spot_details_trading |
| 3207 | games - nft - second - case - plans | 31 | 3207_games_nft_second_case |
| 3208 | thanks - life - sir - company - word | 31 | 3208_thanks_life_sir_company |
| 3209 | protocol - deal - action - home - network | 31 | 3209_protocol_deal_action_home |
| 3210 | position - chart - buy - price - web | 31 | 3210_position_chart_buy_price |
| 3211 | date - minutes - correct - building - hours | 31 | 3211_date_minutes_correct_building |
| 3212 | guy - development - dev - product - break | 31 | 3212_guy_development_dev_product |
| 3213 | earlier - yesterday - strong - opportunity - great | 31 | 3213_earlier_yesterday_strong_opportunity |
| 3214 | ppl - mate - problems - build - tho | 31 | 3214_ppl_mate_problems_build |
| 3215 | couple - minutes - day - hours - new | 31 | 3215_couple_minutes_day_hours |
| 3216 | projects - market - building - project - situation | 31 | 3216_projects_market_building_project |
| 3217 | work - works - previous - luck - shit | 31 | 3217_work_works_previous_luck |
| 3218 | issues - guess - related - thing - country | 31 | 3218_issues_guess_related_thing |
| 3219 | tomorrow - day - good - right - days | 31 | 3219_tomorrow_day_good_right |
| 3220 | latest - official - data - power - phone | 31 | 3220_latest_official_data_power |
| 3221 | feature - projects - building - likely - experience | 31 | 3221_feature_projects_building_likely |
| 3222 | early - benefits - links - power - rewards | 31 | 3222_early_benefits_links_power |
| 3223 | today - total - days - features - hours | 31 | 3223_today_total_days_features |
| 3224 | global - list - official - app - platform | 31 | 3224_global_list_official_app |
| 3225 | site - website - earlier - couple - friend | 31 | 3225_site_website_earlier_couple |
| 3226 | hour - drop - message - service - active | 31 | 3226_hour_drop_message_service |
| 3227 | attention - past - btc - recent - success | 31 | 3227_attention_past_btc_recent |
| 3228 | certain - bitcoin - price - example - weeks | 31 | 3228_certain_bitcoin_price_example |
| 3229 | point - total - profit - report - period | 31 | 3229_point_total_profit_report |
| 3230 | features - mate - financial - assets - dude | 31 | 3230_features_mate_financial_assets |
| 3231 | report - security - members - fake - admin | 31 | 3231_report_security_members_fake |
| 3232 | control - extra - fast - service - legit | 31 | 3232_control_extra_fast_service |
| 3233 | biggest - interesting - things - space - today | 31 | 3233_biggest_interesting_things_space |
| 3234 | launch - ai - hold - low - contract | 31 | 3234_launch_ai_hold_low |
| 3235 | interested - tho - form - main - issues | 31 | 3235_interested_tho_form_main |
| 3236 | old - version - step - clear - new | 31 | 3236_old_version_step_clear |
| 3237 | market - previous - selling - days - spot | 31 | 3237_market_previous_selling_days |
| 3238 | issue - game - fine - site - haha | 31 | 3238_issue_game_fine_site |
| 3239 | - - - - | 31 | 3239____ |
| 3240 | mind - fact - answer - past - sense | 31 | 3240_mind_fact_answer_past |
| 3241 | financial - cryptocurrency - launch - exchange - wallet | 31 | 3241_financial_cryptocurrency_launch_exchange |
| 3242 | money - - - - | 31 | 3242_money___ |
| 3243 | nice - idea - fun - wrong - short | 31 | 3243_nice_idea_fun_wrong |
| 3244 | mate - matter - brother - man - opinion | 31 | 3244_mate_matter_brother_man |
| 3245 | app - guide - ama - video - action | 31 | 3245_app_guide_ama_video |
| 3246 | pls - yesterday - transactions - special - situation | 31 | 3246_pls_yesterday_transactions_special |
| 3247 | rate - huge - ethereum - support - bitcoin | 31 | 3247_rate_huge_ethereum_support |
| 3248 | protocol - potential - future - ama - close | 31 | 3248_protocol_potential_future_ama |
| 3249 | ca - token - answer - question - happy | 31 | 3249_ca_token_answer_question |
| 3250 | previous - main - huge - launch - holders | 31 | 3250_previous_main_huge_launch |
| 3251 | recent - profit - weeks - trading - report | 31 | 3251_recent_profit_weeks_trading |
| 3252 | social - trading - new - - | 31 | 3252_social_trading_new_ |
| 3253 | binance - payment - big - available - | 31 | 3253_binance_payment_big_available |
| 3254 | rest - thanks - email - site - matter | 31 | 3254_rest_thanks_email_site |
| 3255 | fun - dm - bit - available - free | 31 | 3255_fun_dm_bit_available |
| 3256 | ama - product - main - hit - public | 31 | 3256_ama_product_main_hit |
| 3257 | price - correct - certain - wrong - normal | 31 | 3257_price_correct_certain_wrong |
| 3258 | love - online - country - person - problem | 31 | 3258_love_online_country_person |
| 3259 | recent - longer - action - low - btc | 31 | 3259_recent_longer_action_low |
| 3260 | tokens - btc - block - ones - questions | 31 | 3260_tokens_btc_block_ones |
| 3261 | important - things - idea - good - | 31 | 3261_important_things_idea_good |
| 3262 | twitter - projects - crypto - stuff - thing | 31 | 3262_twitter_projects_crypto_stuff |
| 3263 | liquidity - tokens - token - value - problems | 31 | 3263_liquidity_tokens_token_value |
| 3264 | tomorrow - mate - yesterday - brother - today | 31 | 3264_tomorrow_mate_yesterday_brother |
| 3265 | special - family - crazy - kind - hard | 31 | 3265_special_family_crazy_kind |
| 3266 | ai - order - volume - sign - safe | 31 | 3266_ai_order_volume_sign |
| 3267 | recent - bit - morning - reason - times | 31 | 3267_recent_bit_morning_reason |
| 3268 | ai - possible - use - telegram - fact | 31 | 3268_ai_possible_use_telegram |
| 3269 | deal - video - media - potential - access | 31 | 3269_deal_video_media_potential |
| 3270 | live - certain - feature - version - phone | 31 | 3270_live_certain_feature_version |
| 3271 | spot - contact - channel - available - money | 31 | 3271_spot_contact_channel_available |
| 3272 | situation - deal - big - - | 31 | 3272_situation_deal_big_ |
| 3273 | interested - update - site - upcoming - payment | 31 | 3273_interested_update_site_upcoming |
| 3274 | page - day - free - half - dm | 31 | 3274_page_day_free_half |
| 3275 | crypto - set - read - cryptocurrency - | 31 | 3275_crypto_set_read_cryptocurrency |
| 3276 | live - support - reply - site - help | 31 | 3276_live_support_reply_site |
| 3277 | interested - talk - fun - - | 31 | 3277_interested_talk_fun_ |
| 3278 | dude - feature - type - multiple - country | 30 | 3278_dude_feature_type_multiple |
| 3279 | head - special - feature - buddy - better | 30 | 3279_head_special_feature_buddy |
| 3280 | global - launch - late - space - glad | 30 | 3280_global_launch_late_space |
| 3281 | family - years - - - | 30 | 3281_family_years__ |
| 3282 | drop - pool - end - date - updates | 30 | 3282_drop_pool_end_date |
| 3283 | lots - links - high - active - big | 30 | 3283_lots_links_high_active |
| 3284 | selling - sell - assets - crazy - yesterday | 30 | 3284_selling_sell_assets_crazy |
| 3285 | hour - minimum - launch - payment - new | 30 | 3285_hour_minimum_launch_payment |
| 3286 | available - - - - | 30 | 3286_available___ |
| 3287 | tomorrow - way - good - luck - days | 30 | 3287_tomorrow_way_good_luck |
| 3288 | words - kind - dude - thanks - type | 30 | 3288_words_kind_dude_thanks |
| 3289 | data - price - bitcoin - - | 30 | 3289_data_price_bitcoin_ |
| 3290 | idea - project - development - strong - future | 30 | 3290_idea_project_development_strong |
| 3291 | fix - reason - issue - dev - wrong | 30 | 3291_fix_reason_issue_dev |
| 3292 | home - assets - digital - massive - year | 30 | 3292_home_assets_digital_massive |
| 3293 | nft - announcement - claim - reward - form | 30 | 3293_nft_announcement_claim_reward |
| 3294 | - - - - | 30 | 3294____ |
| 3295 | form - crypto - payment - wonder - option | 30 | 3295_form_crypto_payment_wonder |
| 3296 | day - today - nice - night - good | 30 | 3296_day_today_nice_night |
| 3297 | moment - live - current - clear - shit | 30 | 3297_moment_live_current_clear |
| 3298 | admin - group - close - help - true | 30 | 3298_admin_group_close_help |
| 3299 | write - information - private - way - channel | 30 | 3299_write_information_private_way |
| 3300 | risk - level - official - daily - | 30 | 3300_risk_level_official_daily |
| 3301 | protocol - solution - supply - users - platform | 30 | 3301_protocol_solution_supply_users |
| 3302 | bad - guy - nice - good - words | 30 | 3302_bad_guy_nice_good |
| 3303 | step - digital - ready - real - assets | 30 | 3303_step_digital_ready_real |
| 3304 | late - hour - minutes - daily - reward | 30 | 3304_late_hour_minutes_daily |
| 3305 | binance - tokens - protocol - team - situation | 30 | 3305_binance_tokens_protocol_team |
| 3306 | code - withdraw - number - details - issues | 30 | 3306_code_withdraw_number_details |
| 3307 | note - fee - cryptocurrency - fees - address | 30 | 3307_note_fee_cryptocurrency_fees |
| 3308 | dear - sure - - - | 30 | 3308_dear_sure__ |
| 3309 | binance - capital - potential - news - account | 30 | 3309_binance_capital_potential_news |
| 3310 | yesterday - specific - awesome - correct - words | 30 | 3310_yesterday_specific_awesome_correct |
| 3311 | ama - token - higher - mate - fee | 30 | 3311_ama_token_higher_mate |
| 3312 | games - playing - platform - game - account | 30 | 3312_games_playing_platform_game |
| 3313 | code - difference - yesterday - night - country | 30 | 3313_code_difference_yesterday_night |
| 3314 | recent - second - place - market - time | 30 | 3314_recent_second_place_market |
| 3315 | fun - investment - link - - | 30 | 3315_fun_investment_link_ |
| 3316 | reward - quick - ready - opportunity - month | 30 | 3316_reward_quick_ready_opportunity |
| 3317 | spot - message - links - extra - pay | 30 | 3317_spot_message_links_extra |
| 3318 | buy - btc - short - major - long | 30 | 3318_buy_btc_short_major |
| 3319 | minimum - couple - upcoming - option - fine | 30 | 3319_minimum_couple_upcoming_option |
| 3320 | buddy - dude - friend - sir - man | 30 | 3320_buddy_dude_friend_sir |
| 3321 | address - eth - chain - sense - contract | 30 | 3321_address_eth_chain_sense |
| 3322 | bot - ways - free - early - moment | 30 | 3322_bot_ways_free_early |
| 3323 | face - dude - head - yesterday - lots | 30 | 3323_face_dude_head_yesterday |
| 3324 | weeks - form - global - control - product | 30 | 3324_weeks_form_global_control |
| 3325 | bad - good - - - | 30 | 3325_bad_good__ |
| 3326 | likely - higher - ways - yesterday - interesting | 30 | 3326_likely_higher_ways_yesterday |
| 3327 | business - years - binance - wallet - info | 30 | 3327_business_years_binance_wallet |
| 3328 | volume - control - power - key - solution | 30 | 3328_volume_control_power_key |
| 3329 | year - example - members - daily - minimum | 30 | 3329_year_example_members_daily |
| 3330 | home - feature - crypto - lot - likely | 30 | 3330_home_feature_crypto_lot |
| 3331 | minutes - announcement - binance - open - members | 30 | 3331_minutes_announcement_binance_open |
| 3332 | product - clear - guy - cool - fine | 30 | 3332_product_clear_guy_cool |
| 3333 | report - shit - word - massive - higher | 30 | 3333_report_shit_word_massive |
| 3334 | security - product - trust - trade - link | 30 | 3334_security_product_trust_trade |
| 3335 | level - welcome - deposit - new - | 30 | 3335_level_welcome_deposit_new |
| 3336 | interested - real - friends - friend - dear | 30 | 3336_interested_real_friends_friend |
| 3337 | yesterday - bad - good - time - | 30 | 3337_yesterday_bad_good_time |
| 3338 | special - minutes - place - date - chat | 30 | 3338_special_minutes_place_date |
| 3339 | example - crazy - yesterday - stuff - shit | 30 | 3339_example_crazy_yesterday_stuff |
| 3340 | block - report - person - buddy - open | 30 | 3340_block_report_person_buddy |
| 3341 | answer - higher - question - claim - level | 30 | 3341_answer_higher_question_claim |
| 3342 | protocol - ecosystem - network - major - media | 30 | 3342_protocol_ecosystem_network_major |
| 3343 | multiple - tg - marketing - launch - stay | 30 | 3343_multiple_tg_marketing_launch |
| 3344 | fees - holders - liquidity - plan - pool | 30 | 3344_fees_holders_liquidity_plan |
| 3345 | latest - fee - information - future - opportunity | 30 | 3345_latest_fee_information_future |
| 3346 | normal - price - value - couple - number | 30 | 3346_normal_price_value_couple |
| 3347 | latest - hard - hand - online - complete | 30 | 3347_latest_hard_hand_online |
| 3348 | true - love - hours - friend - past | 30 | 3348_true_love_hours_friend |
| 3349 | reply - sir - email - tomorrow - details | 30 | 3349_reply_sir_email_tomorrow |
| 3350 | perfect - plans - business - minimum - online | 30 | 3350_perfect_plans_business_minimum |
| 3351 | lots - links - attention - admin - kind | 30 | 3351_lots_links_attention_admin |
| 3352 | holders - total - - - | 30 | 3352_holders_total__ |
| 3353 | country - phone - users - service - payment | 30 | 3353_country_phone_users_service |
| 3354 | simple - early - chance - launch - contract | 30 | 3354_simple_early_chance_launch |
| 3355 | article - sense - ama - recent - positive | 30 | 3355_article_sense_ama_recent |
| 3356 | transactions - fees - blockchain - high - bitcoin | 30 | 3356_transactions_fees_blockchain_high |
| 3357 | fake - admin - post - updates - official | 30 | 3357_fake_admin_post_updates |
| 3358 | crazy - wonder - dude - bro - tho | 30 | 3358_crazy_wonder_dude_bro |
| 3359 | btc - information - channel - market - link | 30 | 3359_btc_information_channel_market |
| 3360 | kind - easy - face - awesome - luck | 30 | 3360_kind_easy_face_awesome |
| 3361 | binance - complete - user - coins - code | 30 | 3361_binance_complete_user_coins |
| 3362 | blockchain - group - - - | 30 | 3362_blockchain_group__ |
| 3363 | term - running - minimum - short - sign | 30 | 3363_term_running_minimum_short |
| 3364 | deposit - withdraw - half - order - crazy | 30 | 3364_deposit_withdraw_half_order |
| 3365 | person - action - possible - - | 30 | 3365_person_action_possible_ |
| 3366 | technology - power - social - projects - major | 30 | 3366_technology_power_social_projects |
| 3367 | selling - lot - page - active - big | 30 | 3367_selling_lot_page_active |
| 3368 | web - play - discord - announcement - coin | 30 | 3368_web_play_discord_announcement |
| 3369 | bit - problems - family - country - words | 30 | 3369_bit_problems_family_country |
| 3370 | exciting - ecosystem - plans - cryptocurrency - future | 30 | 3370_exciting_ecosystem_plans_cryptocurrency |
| 3371 | participate - pool - information - reward - complete | 30 | 3371_participate_pool_information_reward |
| 3372 | comment - couple - feature - previous - reason | 30 | 3372_comment_couple_feature_previous |
| 3373 | matter - - - - | 30 | 3373_matter___ |
| 3374 | person - hand - web - video - possible | 30 | 3374_person_hand_web_video |
| 3375 | situation - normal - tho - issue - changed | 30 | 3375_situation_normal_tho_issue |
| 3376 | morning - glad - buddy - nice - dear | 30 | 3376_morning_glad_buddy_nice |
| 3377 | rate - assets - protocol - plan - block | 30 | 3377_rate_assets_protocol_plan |
| 3378 | message - link - - - | 30 | 3378_message_link__ |
| 3379 | announcement - discord - information - contact - chat | 30 | 3379_announcement_discord_information_contact |
| 3380 | little - days - - - | 30 | 3380_little_days__ |
| 3381 | report - - - - | 30 | 3381_report___ |
| 3382 | mate - man - country - sir - guys | 30 | 3382_mate_man_country_sir |
| 3383 | trading - crypto - report - assets - problem | 30 | 3383_trading_crypto_report_assets |
| 3384 | successful - trading - profit - happy - trade | 30 | 3384_successful_trading_profit_happy |
| 3385 | problems - global - financial - capital - way | 30 | 3385_problems_global_financial_capital |
| 3386 | spot - idea - short - ta - chart | 30 | 3386_spot_idea_short_ta |
| 3387 | case - year - high - community - bitcoin | 30 | 3387_case_year_high_community |
| 3388 | channel - trading - group - feel - clear | 30 | 3388_channel_trading_group_feel |
| 3389 | data - digital - use - web - community | 30 | 3389_data_digital_use_web |
| 3390 | global - plans - users - market - certain | 30 | 3390_global_plans_users_market |
| 3391 | list - nice - network - - | 30 | 3391_list_nice_network_ |
| 3392 | similar - sorry - late - yesterday - guy | 30 | 3392_similar_sorry_late_yesterday |
| 3393 | win - end - risk - level - official | 30 | 3393_win_end_risk_level |
| 3394 | strong - perfect - community - hand - project | 30 | 3394_strong_perfect_community_hand |
| 3395 | - - - - | 30 | 3395____ |
| 3396 | chat - connect - option - issues - guide | 30 | 3396_chat_connect_option_issues |
| 3397 | coins - tokens - lot - - | 30 | 3397_coins_tokens_lot_ |
| 3398 | chain - services - service - protocol - issue | 30 | 3398_chain_services_service_protocol |
| 3399 | fair - dude - special - opinion - second | 30 | 3399_fair_dude_special_opinion |
| 3400 | short - position - opinion - crazy - yesterday | 30 | 3400_short_position_opinion_crazy |
| 3401 | head - games - minutes - lots - mind | 30 | 3401_head_games_minutes_lots |
| 3402 | investors - trading - night - news - action | 30 | 3402_investors_trading_night_news |
| 3403 | smart - chain - block - network - data | 30 | 3403_smart_chain_block_network |
| 3404 | example - specific - projects - glad - interesting | 30 | 3404_example_specific_projects_glad |
| 3405 | crypto - similar - digital - recent - future | 30 | 3405_crypto_similar_digital_recent |
| 3406 | release - positive - bit - answer - announcement | 30 | 3406_release_positive_bit_answer |
| 3407 | night - games - shit - game - damn | 30 | 3407_night_games_shit_game |
| 3408 | opinion - product - words - personal - amazing | 30 | 3408_opinion_product_words_personal |
| 3409 | deal - order - fact - idea - funds | 30 | 3409_deal_order_fact_idea |
| 3410 | major - awesome - features - launch - huge | 30 | 3410_major_awesome_features_launch |
| 3411 | night - nft - public - drop - crazy | 30 | 3411_night_nft_public_drop |
| 3412 | years - looks - nft - easy - bitcoin | 29 | 3412_years_looks_nft_easy |
| 3413 | great - fine - amazing - hope - thing | 29 | 3413_great_fine_amazing_hope |
| 3414 | answer - related - questions - mate - question | 29 | 3414_answer_related_questions_mate |
| 3415 | development - future - amazing - thing - big | 29 | 3415_development_future_amazing_thing |
| 3416 | weeks - list - true - opportunity - binance | 29 | 3416_weeks_list_true_opportunity |
| 3417 | shit - reply - guy - simple - guess | 29 | 3417_shit_reply_guy_simple |
| 3418 | control - country - opinion - situation - cause | 29 | 3418_control_country_opinion_situation |
| 3419 | market - good - current - nice - luck | 29 | 3419_market_good_current_nice |
| 3420 | form - event - pm - unique - reward | 29 | 3420_form_event_pm_unique |
| 3421 | pretty - version - current - smart - friend | 29 | 3421_pretty_version_current_smart |
| 3422 | control - mind - looks - difference - face | 29 | 3422_control_mind_looks_difference |
| 3423 | ama - web - assets - token - date | 29 | 3423_ama_web_assets_token |
| 3424 | power - version - minutes - strong - family | 29 | 3424_power_version_minutes_strong |
| 3425 | person - benefits - level - media - social | 29 | 3425_person_benefits_level_media |
| 3426 | app - game - key - web - games | 29 | 3426_app_game_key_web |
| 3427 | win - risk - level - official - daily | 29 | 3427_win_risk_level_official |
| 3428 | rules - ai - process - code - world | 29 | 3428_rules_ai_process_code |
| 3429 | network - feature - product - awesome - date | 29 | 3429_network_feature_product_awesome |
| 3430 | hours - new - - - | 29 | 3430_hours_new__ |
| 3431 | days - market - dude - past - single | 29 | 3431_days_market_dude_past |
| 3432 | luck - cool - game - form - win | 29 | 3432_luck_cool_game_form |
| 3433 | service - services - moment - thanks - clear | 29 | 3433_service_services_moment_thanks |
| 3434 | hours - previous - site - phone - discord | 29 | 3434_hours_previous_site_phone |
| 3435 | matter - stuff - special - difference - different | 29 | 3435_matter_stuff_special_difference |
| 3436 | dude - morning - friend - man - sure | 29 | 3436_dude_morning_friend_man |
| 3437 | past - weeks - week - digital - btc | 29 | 3437_past_weeks_week_digital |
| 3438 | accounts - funds - drop - pm - past | 29 | 3438_accounts_funds_drop_pm |
| 3439 | admin - glad - site - run - current | 29 | 3439_admin_glad_site_run |
| 3440 | matter - real - order - fact - idea | 29 | 3440_matter_real_order_fact |
| 3441 | ppl - ones - lol - morning - hard | 29 | 3441_ppl_ones_lol_morning |
| 3442 | option - thing - number - weeks - stuff | 29 | 3442_option_thing_number_weeks |
| 3443 | fake - space - head - media - point | 29 | 3443_fake_space_head_media |
| 3444 | welcome - official - channel - step - questions | 29 | 3444_welcome_official_channel_step |
| 3445 | times - connect - coin - power - fun | 29 | 3445_times_connect_coin_power |
| 3446 | second - right - telegram - free - time | 29 | 3446_second_right_telegram_free |
| 3447 | ready - happy - good - - | 29 | 3447_ready_happy_good_ |
| 3448 | difference - glad - excited - links - care | 29 | 3448_difference_glad_excited_links |
| 3449 | hand - break - luck - fun - sir | 29 | 3449_hand_break_luck_fun |
| 3450 | tho - half - bit - specific - fun | 29 | 3450_tho_half_bit_specific |
| 3451 | year - chat - friend - term - years | 29 | 3451_year_chat_friend_term |
| 3452 | benefits - interested - friends - term - friend | 29 | 3452_benefits_interested_friends_term |
| 3453 | control - total - contract - tokens - | 29 | 3453_control_total_contract_tokens |
| 3454 | person - thinking - mind - true - possible | 29 | 3454_person_thinking_mind_true |
| 3455 | binance - past - cryptocurrency - case - news | 29 | 3455_binance_past_cryptocurrency_case |
| 3456 | super - term - biggest - major - fact | 29 | 3456_super_term_biggest_major |
| 3457 | shit - sense - guy - times - easy | 29 | 3457_shit_sense_guy_times |
| 3458 | man - week - - - | 29 | 3458_man_week__ |
| 3459 | brother - fact - interesting - love - stuff | 29 | 3459_brother_fact_interesting_love |
| 3460 | account - company - business - - | 29 | 3460_account_company_business_ |
| 3461 | course - nft - trading - rest - research | 29 | 3461_course_nft_trading_rest |
| 3462 | long - quick - waiting - today - love | 29 | 3462_long_quick_waiting_today |
| 3463 | short - question - action - btc - price | 29 | 3463_short_question_action_btc |
| 3464 | day - money - business - end - different | 29 | 3464_day_money_business_end |
| 3465 | order - - - - | 29 | 3465_order___ |
| 3466 | higher - details - video - low - information | 29 | 3466_higher_details_video_low |
| 3467 | market - tho - recent - today - spot | 29 | 3467_market_tho_recent_today |
| 3468 | mind - message - important - fake - service | 29 | 3468_mind_message_important_fake |
| 3469 | cause - hand - country - problems - power | 29 | 3469_cause_hand_country_problems |
| 3470 | week - service - - - | 29 | 3470_week_service__ |
| 3471 | bro - thank - thanks - man - brother | 29 | 3471_bro_thank_thanks_man |
| 3472 | early - long - way - days - problems | 29 | 3472_early_long_way_days |
| 3473 | way - big - - - | 29 | 3473_way_big__ |
| 3474 | token - ecosystem - version - coin - chance | 29 | 3474_token_ecosystem_version_coin |
| 3475 | left - specific - special - idea - words | 29 | 3475_left_specific_special_idea |
| 3476 | claim - drop - contract - address - glad | 29 | 3476_claim_drop_contract_address |
| 3477 | claim - code - complete - fee - connect | 29 | 3477_claim_code_complete_fee |
| 3478 | clear - project - development - strong - space | 29 | 3478_clear_project_development_strong |
| 3479 | head - fine - normal - yesterday - bad | 29 | 3479_head_fine_normal_yesterday |
| 3480 | guys - future - better - day - good | 29 | 3480_guys_future_better_day |
| 3481 | interested - previous - happy - best - channel | 29 | 3481_interested_previous_happy_best |
| 3482 | accounts - binance - level - website - support | 29 | 3482_accounts_binance_level_website |
| 3483 | phone - account - shows - likely - note | 29 | 3483_phone_account_shows_likely |
| 3484 | profits - happy - crazy - profit - rate | 29 | 3484_profits_happy_crazy_profit |
| 3485 | global - country - funds - fun - order | 29 | 3485_global_country_funds_fun |
| 3486 | year - fees - worth - blockchain - country | 29 | 3486_year_fees_worth_blockchain |
| 3487 | - - - - | 29 | 3487____ |
| 3488 | situation - home - hand - normal - guy | 29 | 3488_situation_home_hand_normal |
| 3489 | level - welcome - deposit - new - | 29 | 3489_level_welcome_deposit_new |
| 3490 | ones - new - brother - old - need | 29 | 3490_ones_new_brother_old |
| 3491 | half - short - open - big - long | 29 | 3491_half_short_open_big |
| 3492 | guys - strong - man - thanks - chance | 29 | 3492_guys_strong_man_thanks |
| 3493 | period - official - super - total - app | 29 | 3493_period_official_super_total |
| 3494 | similar - near - problems - rest - night | 29 | 3494_similar_near_problems_rest |
| 3495 | ai - interesting - possible - talking - rest | 29 | 3495_ai_interesting_possible_talking |
| 3496 | home - btc - huge - claim - hour | 29 | 3496_home_btc_huge_claim |
| 3497 | clear - list - shit - course - announcement | 29 | 3497_clear_list_shit_course |
| 3498 | rules - financial - place - control - security | 29 | 3498_rules_financial_place_control |
| 3499 | wonder - moment - lot - list - true | 29 | 3499_wonder_moment_lot_list |
| 3500 | deal - - - - | 29 | 3500_deal___ |
| 3501 | fine - thanks - - - | 29 | 3501_fine_thanks__ |
| 3502 | lots - tokens - volume - sure - swap | 29 | 3502_lots_tokens_volume_sure |
| 3503 | damn - normal - glad - shit - country | 29 | 3503_damn_normal_glad_shit |
| 3504 | similar - games - crazy - interesting - stuff | 29 | 3504_similar_games_crazy_interesting |
| 3505 | old - lol - close - years - plan | 29 | 3505_old_lol_close_years |
| 3506 | night - links - media - users - group | 29 | 3506_night_links_media_users |
| 3507 | utc - ecosystem - smart - chain - trade | 29 | 3507_utc_ecosystem_smart_chain |
| 3508 | sense - different - weeks - data - guy | 29 | 3508_sense_different_weeks_data |
| 3509 | specific - technology - token - solution - similar | 29 | 3509_specific_technology_token_solution |
| 3510 | media - social - post - today - | 29 | 3510_media_social_post_today |
| 3511 | country - cryptocurrency - transactions - action - financial | 29 | 3511_country_cryptocurrency_transactions_action |
| 3512 | life - shit - months - true - | 29 | 3512_life_shit_months_true |
| 3513 | yesterday - tomorrow - data - exciting - latest | 29 | 3513_yesterday_tomorrow_data_exciting |
| 3514 | guys - today - lot - great - day | 29 | 3514_guys_today_lot_great |
| 3515 | - - - - | 29 | 3515____ |
| 3516 | kind - feature - reply - attention - man | 29 | 3516_kind_feature_reply_attention |
| 3517 | sorry - buddy - fine - dude - brother | 29 | 3517_sorry_buddy_fine_dude |
| 3518 | country - head - super - likely - problems | 29 | 3518_country_head_super_likely |
| 3519 | ones - ecosystem - shows - fun - space | 29 | 3519_ones_ecosystem_shows_fun |
| 3520 | transaction - fees - thing - trust - marketing | 29 | 3520_transaction_fees_thing_trust |
| 3521 | break - ways - little - fake - easy | 29 | 3521_break_ways_little_fake |
| 3522 | btc - previous - channel - small - article | 29 | 3522_btc_previous_channel_small |
| 3523 | nft - binance - users - bitcoin - platform | 29 | 3523_nft_binance_users_bitcoin |
| 3524 | website - site - previous - new - normal | 29 | 3524_website_site_previous_new |
| 3525 | couple - sense - previous - word - assets | 29 | 3525_couple_sense_previous_word |
| 3526 | block - report - links - funds - admin | 29 | 3526_block_report_links_funds |
| 3527 | similar - biggest - ones - type - course | 29 | 3527_similar_biggest_ones_type |
| 3528 | old - things - love - years - lot | 29 | 3528_old_things_love_years |
| 3529 | specific - opinion - awesome - alot - fine | 29 | 3529_specific_opinion_awesome_alot |
| 3530 | previous - investors - channel - market - telegram | 29 | 3530_previous_investors_channel_market |
| 3531 | guys - dm - man - single - second | 29 | 3531_guys_dm_man_single |
| 3532 | guys - team - successful - potential - launch | 29 | 3532_guys_team_successful_potential |
| 3533 | fake - media - information - real - account | 29 | 3533_fake_media_information_real |
| 3534 | trust - process - hand - mate - special | 29 | 3534_trust_process_hand_mate |
| 3535 | note - free - channel - links - price | 29 | 3535_note_free_channel_links |
| 3536 | ama - supply - secure - app - assets | 29 | 3536_ama_supply_secure_app |
| 3537 | important - biggest - fact - home - idea | 29 | 3537_important_biggest_fact_home |
| 3538 | play - hour - risk - following - level | 29 | 3538_play_hour_risk_following |
| 3539 | moment - ready - mind - sense - rest | 29 | 3539_moment_ready_mind_sense |
| 3540 | single - positive - tomorrow - mind - process | 29 | 3540_single_positive_tomorrow_mind |
| 3541 | - - - - | 29 | 3541____ |
| 3542 | chart - live - trading - - | 29 | 3542_chart_live_trading_ |
| 3543 | specific - true - known - fact - simple | 29 | 3543_specific_true_known_fact |
| 3544 | specific - services - interested - questions - channel | 29 | 3544_specific_services_interested_questions |
| 3545 | admin - guy - dude - shit - type | 29 | 3545_admin_guy_dude_shit |
| 3546 | interested - cryptocurrency - message - course - business | 29 | 3546_interested_cryptocurrency_message_course |
| 3547 | sure - certain - way - life - right | 29 | 3547_sure_certain_way_life |
| 3548 | security - update - latest - release - specific | 29 | 3548_security_update_latest_release |
| 3549 | clear - focus - half - moment - main | 29 | 3549_clear_focus_half_moment |
| 3550 | announcement - buddy - official - date - update | 29 | 3550_announcement_buddy_official_date |
| 3551 | things - week - bitcoin - current - imagine | 29 | 3551_things_week_bitcoin_current |
| 3552 | option - open - couple - head - contract | 29 | 3552_option_open_couple_head |
| 3553 | network - unique - current - website - fun | 29 | 3553_network_unique_current_website |
| 3554 | idea - man - point - old - things | 28 | 3554_idea_man_point_old |
| 3555 | country - available - interesting - form - use | 28 | 3555_country_available_interesting_form |
| 3556 | low - person - ppl - certain - higher | 28 | 3556_low_person_ppl_certain |
| 3557 | order - wallet - option - link - people | 28 | 3557_order_wallet_option_link |
| 3558 | - - - - | 28 | 3558____ |
| 3559 | awesome - certain - brother - product - stuff | 28 | 3559_awesome_certain_brother_product |
| 3560 | guy - lot - thing - sure - available | 28 | 3560_guy_lot_thing_sure |
| 3561 | discord - details - listing - official - website | 28 | 3561_discord_details_listing_official |
| 3562 | buddy - mate - sure - great - certain | 28 | 3562_buddy_mate_sure_great |
| 3563 | buddy - glad - normal - mate - bro | 28 | 3563_buddy_glad_normal_mate |
| 3564 | block - global - ama - launch - fee | 28 | 3564_block_global_ama_launch |
| 3565 | action - binance - case - - | 28 | 3565_action_binance_case_ |
| 3566 | global - action - financial - money - | 28 | 3566_global_action_financial_money |
| 3567 | great - nice - awesome - recent - luck | 28 | 3567_great_nice_awesome_recent |
| 3568 | control - funds - tokens - security - | 28 | 3568_control_funds_tokens_security |
| 3569 | stuff - rest - night - better - idea | 28 | 3569_stuff_rest_night_better |
| 3570 | message - - - - | 28 | 3570_message___ |
| 3571 | digital - binance - exchange - new - business | 28 | 3571_digital_binance_exchange_new |
| 3572 | certain - sorry - thing - - | 28 | 3572_certain_sorry_thing_ |
| 3573 | updates - update - latest - project - information | 28 | 3573_updates_update_latest_project |
| 3574 | position - exchange - exciting - assets - trade | 28 | 3574_position_exchange_exciting_assets |
| 3575 | code - months - data - - | 28 | 3575_code_months_data_ |
| 3576 | team - likely - funds - pay - control | 28 | 3576_team_likely_funds_pay |
| 3577 | short - open - correct - reply - weeks | 28 | 3577_short_open_correct_reply |
| 3578 | new - market - - - | 28 | 3578_new_market__ |
| 3579 | exciting - rewards - welcome - come - community | 28 | 3579_exciting_rewards_welcome_come |
| 3580 | glad - true - nice - bro - happy | 28 | 3580_glad_true_nice_bro |
| 3581 | nice - space - release - correct - kind | 28 | 3581_nice_space_release_correct |
| 3582 | morning - fun - interested - dear - private | 28 | 3582_morning_fun_interested_dear |
| 3583 | transactions - bit - benefits - wrong - holders | 28 | 3583_transactions_bit_benefits_wrong |
| 3584 | rest - break - problems - week - worth | 28 | 3584_rest_break_problems_week |
| 3585 | deal - wrong - business - investing - lot | 28 | 3585_deal_wrong_business_investing |
| 3586 | - - - - | 28 | 3586____ |
| 3587 | weeks - assets - details - worth - media | 28 | 3587_weeks_assets_details_worth |
| 3588 | coins - main - email - worth - open | 28 | 3588_coins_main_email_worth |
| 3589 | moment - mind - important - fake - accounts | 28 | 3589_moment_mind_important_fake |
| 3590 | bitcoin - excited - country - rewards - app | 28 | 3590_bitcoin_excited_country_rewards |
| 3591 | previous - massive - marketing - number - assets | 28 | 3591_previous_massive_marketing_number |
| 3592 | daily - - - - | 28 | 3592_daily___ |
| 3593 | market - - - - | 28 | 3593_market___ |
| 3594 | enjoy - financial - platform - happy - right | 28 | 3594_enjoy_financial_platform_happy |
| 3595 | couple - single - chat - fun - interested | 28 | 3595_couple_single_chat_fun |
| 3596 | major - space - news - volume - latest | 28 | 3596_major_space_news_volume |
| 3597 | term - true - ones - word - case | 28 | 3597_term_true_ones_word |
| 3598 | funds - withdraw - account - problem - control | 28 | 3598_funds_withdraw_account_problem |
| 3599 | dude - likely - number - yesterday - alot | 28 | 3599_dude_likely_number_yesterday |
| 3600 | doubt - safe - online - happy - life | 28 | 3600_doubt_safe_online_happy |
| 3601 | fun - night - dm - ready - time | 28 | 3601_fun_night_dm_ready |
| 3602 | reply - guys - thanks - trading - | 28 | 3602_reply_guys_thanks_trading |
| 3603 | dude - reason - awesome - past - word | 28 | 3603_dude_reason_awesome_past |
| 3604 | - - - - | 28 | 3604____ |
| 3605 | yesterday - saying - service - money - platform | 28 | 3605_yesterday_saying_service_money |
| 3606 | country - use - crazy - shit - guy | 28 | 3606_country_use_crazy_shit |
| 3607 | strong - time - - - | 28 | 3607_strong_time__ |
| 3608 | accounts - site - personal - business - buy | 28 | 3608_accounts_site_personal_business |
| 3609 | dude - man - guy - head - fact | 28 | 3609_dude_man_guy_head |
| 3610 | run - question - ai - rate - network | 28 | 3610_run_question_ai_rate |
| 3611 | market - run - article - control - recent | 28 | 3611_market_run_article_control |
| 3612 | ama - sir - space - post - video | 28 | 3612_ama_sir_space_post |
| 3613 | risk - level - official - daily - | 28 | 3613_risk_level_official_daily |
| 3614 | pay - binance - admin - minimum - hours | 28 | 3614_pay_binance_admin_minimum |
| 3615 | ai - bot - data - development - awesome | 28 | 3615_ai_bot_data_development |
| 3616 | hand - night - specific - sense - imagine | 28 | 3616_hand_night_specific_sense |
| 3617 | global - users - project - community - different | 28 | 3617_global_users_project_community |
| 3618 | kinda - hand - dude - step - correct | 28 | 3618_kinda_hand_dude_step |
| 3619 | example - correct - simple - rewards - later | 28 | 3619_example_correct_simple_rewards |
| 3620 | value - times - business - face - product | 28 | 3620_value_times_business_face |
| 3621 | great - fine - little - years - group | 28 | 3621_great_fine_little_years |
| 3622 | world - people - good - - | 28 | 3622_world_people_good_ |
| 3623 | moment - massive - doubt - little - hold | 28 | 3623_moment_massive_doubt_little |
| 3624 | questions - question - answer - details - head | 28 | 3624_questions_question_answer_details |
| 3625 | blockchain - crypto - - - | 28 | 3625_blockchain_crypto__ |
| 3626 | successful - person - world - success - end | 28 | 3626_successful_person_world_success |
| 3627 | benefits - care - profits - level - chat | 28 | 3627_benefits_care_profits_level |
| 3628 | digital - assets - event - building - blockchain | 28 | 3628_digital_assets_event_building |
| 3629 | certain - problems - reason - ways - lot | 28 | 3629_certain_problems_reason_ways |
| 3630 | couple - point - stuff - shit - guy | 28 | 3630_couple_point_stuff_shit |
| 3631 | tho - ones - bot - fun - users | 28 | 3631_tho_ones_bot_fun |
| 3632 | price - latest - news - dear - event | 28 | 3632_price_latest_news_dear |
| 3633 | digital - financial - success - website - investment | 28 | 3633_digital_financial_success_website |
| 3634 | hope - best - tho - option - better | 28 | 3634_hope_best_tho_option |
| 3635 | friends - discord - event - period - interested | 28 | 3635_friends_discord_event_period |
| 3636 | link - - - - | 28 | 3636_link___ |
| 3637 | page - - - - | 28 | 3637_page___ |
| 3638 | weeks - complete - simple - period - hours | 28 | 3638_weeks_complete_simple_period |
| 3639 | - - - - | 28 | 3639____ |
| 3640 | report - bitcoin - deal - head - research | 28 | 3640_report_bitcoin_deal_head |
| 3641 | upcoming - digital - global - development - project | 28 | 3641_upcoming_digital_global_development |
| 3642 | blockchain - crypto - likely - protocol - problems | 28 | 3642_blockchain_crypto_likely_protocol |
| 3643 | situation - links - solution - love - stuff | 28 | 3643_situation_links_solution_love |
| 3644 | interesting - gone - head - lots - case | 28 | 3644_interesting_gone_head_lots |
| 3645 | worth - projects - market - - | 28 | 3645_worth_projects_market_ |
| 3646 | biggest - question - dev - shit - word | 28 | 3646_biggest_question_dev_shit |
| 3647 | risk - level - official - daily - | 28 | 3647_risk_level_official_daily |
| 3648 | run - major - cool - hold - man | 28 | 3648_run_major_cool_hold |
| 3649 | fees - volume - months - user - investors | 28 | 3649_fees_volume_months_user |
| 3650 | similar - situation - short - face - buddy | 28 | 3650_similar_situation_short_face |
| 3651 | rewards - great - event - exciting - luck | 28 | 3651_rewards_great_event_exciting |
| 3652 | fair - supply - low - launch - network | 28 | 3652_fair_supply_low_launch |
| 3653 | word - option - guys - information - way | 28 | 3653_word_option_guys_information |
| 3654 | details - news - channel - account - exciting | 28 | 3654_details_news_channel_account |
| 3655 | likely - option - specific - longer - trading | 28 | 3655_likely_option_specific_longer |
| 3656 | discord - links - attention - near - legit | 28 | 3656_discord_links_attention_near |
| 3657 | issues - minutes - crypto - news - services | 28 | 3657_issues_minutes_crypto_news |
| 3658 | global - world - funds - app - crypto | 28 | 3658_global_world_funds_app |
| 3659 | development - wallet - product - protocol - company | 28 | 3659_development_wallet_product_protocol |
| 3660 | game - code - announcement - data - issue | 28 | 3660_game_code_announcement_data |
| 3661 | ready - bitcoin - - - | 28 | 3661_ready_bitcoin__ |
| 3662 | example - hard - work - interesting - brother | 28 | 3662_example_hard_work_interesting |
| 3663 | issues - business - number - plans - crypto | 28 | 3663_issues_business_number_plans |
| 3664 | rewards - place - end - game - day | 28 | 3664_rewards_place_end_game |
| 3665 | head - hand - damn - unique - biggest | 28 | 3665_head_hand_damn_unique |
| 3666 | profit - feature - normal - interesting - change | 28 | 3666_profit_feature_normal_interesting |
| 3667 | website - article - guy - join - person | 28 | 3667_website_article_guy_join |
| 3668 | guys - - - - | 28 | 3668_guys___ |
| 3669 | family - home - situation - guy - country | 28 | 3669_family_home_situation_guy |
| 3670 | word - example - tho - cool - use | 28 | 3670_word_example_tho_cool |
| 3671 | clear - solution - hour - kind - key | 28 | 3671_clear_solution_hour_kind |
| 3672 | successful - date - payment - easy - access | 28 | 3672_successful_date_payment_easy |
| 3673 | difference - specific - happening - morning - normal | 28 | 3673_difference_specific_happening_morning |
| 3674 | token - worth - tokens - type - mind | 28 | 3674_token_worth_tokens_type |
| 3675 | details - ethereum - bot - address - man | 28 | 3675_details_ethereum_bot_address |
| 3676 | data - transactions - possible - blockchain - community | 28 | 3676_data_transactions_possible_blockchain |
| 3677 | half - option - hour - bit - idea | 28 | 3677_half_option_hour_bit |
| 3678 | reason - project - - - | 28 | 3678_reason_project__ |
| 3679 | cool - country - place - nice - dear | 28 | 3679_cool_country_place_nice |
| 3680 | project - couple - idea - certain - head | 28 | 3680_project_couple_idea_certain |
| 3681 | coins - visit - interested - worth - person | 28 | 3681_coins_visit_interested_worth |
| 3682 | stuff - shit - kind - things - bro | 28 | 3682_stuff_shit_kind_things |
| 3683 | bitcoin - - - - | 28 | 3683_bitcoin___ |
| 3684 | ones - dude - things - example - question | 28 | 3684_ones_dude_things_example |
| 3685 | dude - yesterday - pretty - true - ready | 28 | 3685_dude_yesterday_pretty_true |
| 3686 | difference - fact - recent - alot - plus | 28 | 3686_difference_fact_recent_alot |
| 3687 | good - morning - day - mate - interesting | 28 | 3687_good_morning_day_mate |
| 3688 | bit - holders - process - token - near | 28 | 3688_bit_holders_process_token |
| 3689 | binance - ones - okay - volume - close | 28 | 3689_binance_ones_okay_volume |
| 3690 | cause - kinda - person - night - things | 28 | 3690_cause_kinda_person_night |
| 3691 | simple - hand - opinion - interesting - stuff | 28 | 3691_simple_hand_opinion_interesting |
| 3692 | deposit - daily - chance - example - sorry | 28 | 3692_deposit_daily_chance_example |
| 3693 | - - - - | 28 | 3693____ |
| 3694 | night - users - end - group - example | 28 | 3694_night_users_end_group |
| 3695 | deal - admin - available - real - company | 28 | 3695_deal_admin_available_real |
| 3696 | ecosystem - little - total - members - app | 28 | 3696_ecosystem_little_total_members |
| 3697 | type - data - complete - shit - success | 28 | 3697_type_data_complete_shit |
| 3698 | change - night - quick - tomorrow - thanks | 27 | 3698_change_night_quick_tomorrow |
| 3699 | question - btc - trading - single - personal | 27 | 3699_question_btc_trading_single |
| 3700 | news - certain - thought - good - question | 27 | 3700_news_certain_thought_good |
| 3701 | support - coins - lot - - | 27 | 3701_support_coins_lot_ |
| 3702 | cool - awesome - guy - simple - things | 27 | 3702_cool_awesome_guy_simple |
| 3703 | situation - kinda - best - opinion - better | 27 | 3703_situation_kinda_best_opinion |
| 3704 | link - - - - | 27 | 3704_link___ |
| 3705 | minimum - profit - plan - fee - days | 27 | 3705_minimum_profit_plan_fee |
| 3706 | place - usdt - bot - date - pool | 27 | 3706_place_usdt_bot_date |
| 3707 | features - smart - massive - supply - total | 27 | 3707_features_smart_massive_supply |
| 3708 | problem - chain - product - chance - crazy | 27 | 3708_problem_chain_product_chance |
| 3709 | tokens - certain - volume - token - | 27 | 3709_tokens_certain_volume_token |
| 3710 | - - - - | 27 | 3710____ |
| 3711 | opinion - certain - fair - public - questions | 27 | 3711_opinion_certain_fair_public |
| 3712 | rules - questions - welcome - community - chat | 27 | 3712_rules_questions_welcome_community |
| 3713 | list - little - safe - private - try | 27 | 3713_list_little_safe_private |
| 3714 | fact - positive - things - thing - case | 27 | 3714_fact_positive_things_thing |
| 3715 | - - - - | 27 | 3715____ |
| 3716 | point - clear - past - shows - moment | 27 | 3716_point_clear_past_shows |
| 3717 | cool - day - account - - | 27 | 3717_cool_day_account_ |
| 3718 | interested - possible - person - project - chain | 27 | 3718_interested_possible_person_project |
| 3719 | updates - channel - order - payment - rate | 27 | 3719_updates_channel_order_payment |
| 3720 | day - life - best - days - good | 27 | 3720_day_life_best_days |
| 3721 | swap - blockchain - fees - word - exciting | 27 | 3721_swap_blockchain_fees_word |
| 3722 | sell - account - business - available - days | 27 | 3722_sell_account_business_available |
| 3723 | world - official - website - awesome - ama | 27 | 3723_world_official_website_awesome |
| 3724 | chart - trade - market - new - position | 27 | 3724_chart_trade_market_new |
| 3725 | worth - man - investment - - | 27 | 3725_worth_man_investment_ |
| 3726 | tomorrow - pm - pay - money - luck | 27 | 3726_tomorrow_pm_pay_money |
| 3727 | visit - months - rate - information - work | 27 | 3727_visit_months_rate_information |
| 3728 | luck - exciting - updates - benefits - total | 27 | 3728_luck_exciting_updates_benefits |
| 3729 | article - dev - main - fact - moment | 27 | 3729_article_dev_main_fact |
| 3730 | position - chart - holders - funds - buy | 27 | 3730_position_chart_holders_funds |
| 3731 | late - special - live - issues - unique | 27 | 3731_late_special_live_issues |
| 3732 | couple - guy - members - guys - months | 27 | 3732_couple_guy_members_guys |
| 3733 | free - - - - | 27 | 3733_free___ |
| 3734 | easy - business - things - opportunity - good | 27 | 3734_easy_business_things_opportunity |
| 3735 | volume - swap - price - buy - market | 27 | 3735_volume_swap_price_buy |
| 3736 | issue - special - solution - project - shit | 27 | 3736_issue_special_solution_project |
| 3737 | late - form - payment - security - funds | 27 | 3737_late_form_payment_security |
| 3738 | - - - - | 27 | 3738____ |
| 3739 | email - hello - contact - dear - free | 27 | 3739_email_hello_contact_dear |
| 3740 | clear - ecosystem - end - year - set | 27 | 3740_clear_ecosystem_end_year |
| 3741 | platform - chance - - - | 27 | 3741_platform_chance__ |
| 3742 | message - personal - - - | 27 | 3742_message_personal__ |
| 3743 | phone - smart - list - contact - easy | 27 | 3743_phone_smart_list_contact |
| 3744 | - - - - | 27 | 3744____ |
| 3745 | recent - search - rules - short - access | 27 | 3745_recent_search_rules_short |
| 3746 | ecosystem - data - interested - code - blockchain | 27 | 3746_ecosystem_data_interested_code |
| 3747 | world - key - action - better - true | 27 | 3747_world_key_action_better |
| 3748 | close - dude - problems - wrong - huge | 27 | 3748_close_dude_problems_wrong |
| 3749 | reason - upcoming - exciting - game - event | 27 | 3749_reason_upcoming_exciting_game |
| 3750 | successful - information - people - - | 27 | 3750_successful_information_people_ |
| 3751 | contract - token - - - | 27 | 3751_contract_token__ |
| 3752 | protocol - defi - early - minimum - special | 27 | 3752_protocol_defi_early_minimum |
| 3753 | problem - mate - situation - buddy - sir | 27 | 3753_problem_mate_situation_buddy |
| 3754 | deposit - alright - funds - order - normal | 27 | 3754_deposit_alright_funds_order |
| 3755 | cause - difference - try - wrong - order | 27 | 3755_cause_difference_try_wrong |
| 3756 | minimum - withdrawal - withdraw - sir - fee | 27 | 3756_minimum_withdrawal_withdraw_sir |
| 3757 | exchange - financial - - - | 27 | 3757_exchange_financial__ |
| 3758 | point - correct - difference - mate - strong | 27 | 3758_point_correct_difference_mate |
| 3759 | chance - change - financial - life - help | 27 | 3759_chance_change_financial_life |
| 3760 | dude - stuff - moment - guy - man | 27 | 3760_dude_stuff_moment_guy |
| 3761 | win - risk - level - official - daily | 27 | 3761_win_risk_level_official |
| 3762 | level - claim - free - hours - payment | 27 | 3762_level_claim_free_hours |
| 3763 | projects - term - coin - situation - coins | 27 | 3763_projects_term_coin_situation |
| 3764 | control - ecosystem - rules - marketing - world | 27 | 3764_control_ecosystem_rules_marketing |
| 3765 | interested - legit - daily - huge - profit | 27 | 3765_interested_legit_daily_huge |
| 3766 | special - assets - moment - ones - game | 27 | 3766_special_assets_moment_ones |
| 3767 | biggest - solution - ethereum - coins - network | 27 | 3767_biggest_solution_ethereum_coins |
| 3768 | face - ones - specific - reply - sorry | 27 | 3768_face_ones_specific_reply |
| 3769 | global - welcome - community - website - money | 27 | 3769_global_welcome_community_website |
| 3770 | bot - feature - message - reply - user | 27 | 3770_bot_feature_message_reply |
| 3771 | shit - rest - country - spot - main | 27 | 3771_shit_rest_country_spot |
| 3772 | hand - months - product - example - business | 27 | 3772_hand_months_product_example |
| 3773 | opinion - hand - fact - interesting - guy | 27 | 3773_opinion_hand_fact_interesting |
| 3774 | product - stuff - ta - job - lol | 27 | 3774_product_stuff_ta_job |
| 3775 | launch - global - ama - contract - week | 27 | 3775_launch_global_ama_contract |
| 3776 | amazing - offer - account - available - deal | 27 | 3776_amazing_offer_account_available |
| 3777 | holders - links - exchange - swap - supply | 27 | 3777_holders_links_exchange_swap |
| 3778 | crazy - huge - ready - week - month | 27 | 3778_crazy_huge_ready_week |
| 3779 | fact - positive - mind - work - possible | 27 | 3779_fact_positive_mind_work |
| 3780 | features - financial - lots - potential - success | 27 | 3780_features_financial_lots_potential |
| 3781 | app - network - project - digital - simple | 27 | 3781_app_network_project_digital |
| 3782 | half - normal - major - weeks - option | 27 | 3782_half_normal_major_weeks |
| 3783 | list - interesting - focus - single - main | 27 | 3783_list_interesting_focus_single |
| 3784 | current - price - new - - | 27 | 3784_current_price_new_ |
| 3785 | buddy - man - dude - bro - guy | 27 | 3785_buddy_man_dude_bro |
| 3786 | damn - mind - glad - crazy - guy | 27 | 3786_damn_mind_glad_crazy |
| 3787 | guide - cryptocurrency - complete - trading - alot | 27 | 3787_guide_cryptocurrency_complete_trading |
| 3788 | previous - step - experience - mind - opportunity | 27 | 3788_previous_step_experience_mind |
| 3789 | reason - bro - better - - | 27 | 3789_reason_bro_better_ |
| 3790 | media - open - social - message - | 27 | 3790_media_open_social_message |
| 3791 | risk - biggest - profit - trading - worth | 27 | 3791_risk_biggest_profit_trading |
| 3792 | updates - building - step - chance - love | 27 | 3792_updates_building_step_chance |
| 3793 | half - supply - fix - increase - clear | 27 | 3793_half_supply_fix_increase |
| 3794 | channel - telegram - risk - man - hours | 27 | 3794_channel_telegram_risk_man |
| 3795 | buddy - times - lot - thanks - platform | 27 | 3795_buddy_times_lot_thanks |
| 3796 | certain - clear - questions - yesterday - simple | 27 | 3796_certain_clear_questions_yesterday |
| 3797 | luck - mate - - - | 27 | 3797_luck_mate__ |
| 3798 | form - reply - term - ways - shows | 27 | 3798_form_reply_term_ways |
| 3799 | message - big - today - - | 27 | 3799_message_big_today_ |
| 3800 | private - smart - product - solution - protocol | 27 | 3800_private_smart_product_solution |
| 3801 | usdt - message - lot - bit - pm | 27 | 3801_usdt_message_lot_bit |
| 3802 | country - free - half - rest - person | 27 | 3802_country_free_half_rest |
| 3803 | chain - announcement - dear - details - normal | 27 | 3803_chain_announcement_dear_details |
| 3804 | glad - form - reason - things - better | 27 | 3804_glad_form_reason_things |
| 3805 | chat - question - example - solution - recent | 27 | 3805_chat_question_example_solution |
| 3806 | position - exchange - business - marketing - community | 27 | 3806_position_exchange_business_marketing |
| 3807 | phone - issue - problem - latest - normal | 27 | 3807_phone_issue_problem_latest |
| 3808 | code - trade - worth - minimum - address | 27 | 3808_code_trade_worth_minimum |
| 3809 | details - correct - plus - guy - transactions | 27 | 3809_details_correct_plus_guy |
| 3810 | public - private - sense - article - stuff | 27 | 3810_public_private_sense_article |
| 3811 | article - fact - spot - words - different | 27 | 3811_article_fact_spot_words |
| 3812 | shit - dude - sense - article - normal | 27 | 3812_shit_dude_sense_article |
| 3813 | details - message - account - support - available | 27 | 3813_details_message_account_support |
| 3814 | days - new - projects - week - | 27 | 3814_days_new_projects_week |
| 3815 | block - phone - awesome - bit - guide | 27 | 3815_block_phone_awesome_bit |
| 3816 | pm - hour - day - small - power | 27 | 3816_pm_hour_day_small |
| 3817 | easy - turn - daily - site - features | 27 | 3817_easy_turn_daily_site |
| 3818 | main - trust - safe - chat - ways | 27 | 3818_main_trust_safe_chat |
| 3819 | shit - similar - brother - development - past | 27 | 3819_shit_similar_brother_development |
| 3820 | bit - mate - lol - man - real | 27 | 3820_bit_mate_lol_man |
| 3821 | phone - later - normal - crazy - issue | 27 | 3821_phone_later_normal_crazy |
| 3822 | email - free - token - rate - days | 27 | 3822_email_free_token_rate |
| 3823 | company - - - - | 27 | 3823_company___ |
| 3824 | chat - way - good - - | 27 | 3824_chat_way_good_ |
| 3825 | couple - fact - positive - job - lots | 27 | 3825_couple_fact_positive_job |
| 3826 | half - case - years - major - recent | 27 | 3826_half_case_years_major |
| 3827 | guys - weeks - past - man - | 27 | 3827_guys_weeks_past_man |
| 3828 | weeks - week - year - idea - sure | 27 | 3828_weeks_week_year_idea |
| 3829 | low - issues - sounds - problems - volume | 27 | 3829_low_issues_sounds_problems |
| 3830 | exchange - bitcoin - - - | 27 | 3830_exchange_bitcoin__ |
| 3831 | user - personal - different - public - links | 27 | 3831_user_personal_different_public |
| 3832 | message - page - great - - | 27 | 3832_message_page_great_ |
| 3833 | action - post - channel - message - | 27 | 3833_action_post_channel_message |
| 3834 | today - work - - - | 27 | 3834_today_work__ |
| 3835 | assets - real - project - interested - message | 26 | 3835_assets_real_project_interested |
| 3836 | brother - product - case - digital - trading | 26 | 3836_brother_product_case_digital |
| 3837 | listing - game - binance - - | 26 | 3837_listing_game_binance_ |
| 3838 | ethereum - years - wallet - worth - eth | 26 | 3838_ethereum_years_wallet_worth |
| 3839 | luck - investing - thing - market - time | 26 | 3839_luck_investing_thing_market |
| 3840 | huge - public - previous - major - upcoming | 26 | 3840_huge_public_previous_major |
| 3841 | situation - upcoming - focus - current - unique | 26 | 3841_situation_upcoming_focus_current |
| 3842 | technology - excited - unique - development - blockchain | 26 | 3842_technology_excited_unique_development |
| 3843 | risk - chain - news - issue - accounts | 26 | 3843_risk_chain_news_issue |
| 3844 | key - service - feature - product - yesterday | 26 | 3844_key_service_feature_product |
| 3845 | single - video - thing - available - time | 26 | 3845_single_video_thing_available |
| 3846 | similar - data - major - recent - exchange | 26 | 3846_similar_data_major_recent |
| 3847 | strong - likely - head - difference - stuff | 26 | 3847_strong_likely_head_difference |
| 3848 | product - defi - user - fact - level | 26 | 3848_product_defi_user_fact |
| 3849 | available - - - - | 26 | 3849_available___ |
| 3850 | example - thinking - supply - similar - year | 26 | 3850_example_thinking_supply_similar |
| 3851 | links - information - morning - official - sure | 26 | 3851_links_information_morning_official |
| 3852 | set - end - - - | 26 | 3852_set_end__ |
| 3853 | simple - mind - happy - moment - main | 26 | 3853_simple_mind_happy_moment |
| 3854 | brother - cryptocurrency - thanks - day - bitcoin | 26 | 3854_brother_cryptocurrency_thanks_day |
| 3855 | certain - interesting - research - earlier - rewards | 26 | 3855_certain_interesting_research_earlier |
| 3856 | selling - binance - bitcoin - country - issues | 26 | 3856_selling_binance_bitcoin_country |
| 3857 | - - - - | 26 | 3857____ |
| 3858 | win - risk - level - official - daily | 26 | 3858_win_risk_level_official |
| 3859 | deal - wrong - business - investing - lot | 26 | 3859_deal_wrong_business_investing |
| 3860 | exchange - - - - | 26 | 3860_exchange___ |
| 3861 | face - shit - building - fake - unique | 26 | 3861_face_shit_building_fake |
| 3862 | private - block - official - admin - bot | 26 | 3862_private_block_official_admin |
| 3863 | stuff - year - - - | 26 | 3863_stuff_year__ |
| 3864 | old - new - - - | 26 | 3864_old_new__ |
| 3865 | country - personal - opinion - guy - bit | 26 | 3865_country_personal_opinion_guy |
| 3866 | luck - feature - late - web - site | 26 | 3866_luck_feature_late_web |
| 3867 | job - certain - example - similar - article | 26 | 3867_job_certain_example_similar |
| 3868 | shit - man - long - time - | 26 | 3868_shit_man_long_time |
| 3869 | google - feature - recent - option - experience | 26 | 3869_google_feature_recent_option |
| 3870 | coins - claim - ecosystem - ethereum - chance | 26 | 3870_coins_claim_ecosystem_ethereum |
| 3871 | network - fees - team - deal - blockchain | 26 | 3871_network_fees_team_deal |
| 3872 | talk - bitcoin - investors - chart - reach | 26 | 3872_talk_bitcoin_investors_chart |
| 3873 | feature - excited - guide - app - old | 26 | 3873_feature_excited_guide_app |
| 3874 | minutes - ethereum - transactions - hour - normal | 26 | 3874_minutes_ethereum_transactions_hour |
| 3875 | idea - question - - - | 26 | 3875_idea_question__ |
| 3876 | week - - - - | 26 | 3876_week___ |
| 3877 | country - public - ai - - | 26 | 3877_country_public_ai_ |
| 3878 | details - bad - way - website - info | 26 | 3878_details_bad_way_website |
| 3879 | wallet - - - - | 26 | 3879_wallet___ |
| 3880 | position - blockchain - team - experience - members | 26 | 3880_position_blockchain_team_experience |
| 3881 | previous - upcoming - attention - massive - assets | 26 | 3881_previous_upcoming_attention_massive |
| 3882 | games - fact - run - term - product | 26 | 3882_games_fact_run_term |
| 3883 | best - - - - | 26 | 3883_best___ |
| 3884 | connect - holders - guide - supply - tokens | 26 | 3884_connect_holders_guide_supply |
| 3885 | cause - action - price - market - | 26 | 3885_cause_action_price_market |
| 3886 | clear - words - excited - platform - real | 26 | 3886_clear_words_excited_platform |
| 3887 | article - guys - good - mate - talk | 26 | 3887_article_guys_good_mate |
| 3888 | previous - weeks - ones - chart - question | 26 | 3888_previous_weeks_ones_chart |
| 3889 | trade - - - - | 26 | 3889_trade___ |
| 3890 | app - video - stuff - feature - latest | 26 | 3890_app_video_stuff_feature |
| 3891 | digital - deal - control - solution - past | 26 | 3891_digital_deal_control_solution |
| 3892 | message - value - marketing - contact - today | 26 | 3892_message_value_marketing_contact |
| 3893 | interested - information - group - latest - rate | 26 | 3893_interested_information_group_latest |
| 3894 | event - man - friend - live - country | 26 | 3894_event_man_friend_live |
| 3895 | version - biggest - major - interesting - unique | 26 | 3895_version_biggest_major_interesting |
| 3896 | massive - single - year - value - funds | 26 | 3896_massive_single_year_value |
| 3897 | situation - guys - doing - better - good | 26 | 3897_situation_guys_doing_better |
| 3898 | interested - video - similar - difference - channel | 26 | 3898_interested_video_similar_difference |
| 3899 | swap - fees - question - plans - transactions | 26 | 3899_swap_fees_question_plans |
| 3900 | level - way - - - | 26 | 3900_level_way__ |
| 3901 | projects - project - low - price - opinion | 26 | 3901_projects_project_low_price |
| 3902 | drop - hour - day - good - dm | 26 | 3902_drop_hour_day_good |
| 3903 | real - form - coins - - | 26 | 3903_real_form_coins_ |
| 3904 | account - accounts - reply - reason - half | 26 | 3904_account_accounts_reply_reason |
| 3905 | assets - value - - - | 26 | 3905_assets_value__ |
| 3906 | dude - old - smart - bad - mean | 26 | 3906_dude_old_smart_bad |
| 3907 | opinion - normal - clear - correct - single | 26 | 3907_opinion_normal_clear_correct |
| 3908 | volume - rewards - current - change - address | 26 | 3908_volume_rewards_current_change |
| 3909 | country - alright - time - nice - friend | 26 | 3909_country_alright_time_nice |
| 3910 | latest - date - updates - media - social | 26 | 3910_latest_date_updates_media |
| 3911 | note - page - active - fast - available | 26 | 3911_note_page_active_fast |
| 3912 | small - end - games - step - difference | 26 | 3912_small_end_games_step |
| 3913 | strong - hand - problems - head - man | 26 | 3913_strong_hand_problems_head |
| 3914 | special - video - available - tomorrow - short | 26 | 3914_special_video_available_tomorrow |
| 3915 | stuff - hour - things - thanks - guys | 26 | 3915_stuff_hour_things_thanks |
| 3916 | event - complete - reward - happy - wallet | 26 | 3916_event_complete_reward_happy |
| 3917 | action - link - mind - - | 26 | 3917_action_link_mind_ |
| 3918 | home - job - friend - months - bad | 26 | 3918_home_job_friend_months |
| 3919 | lots - couple - crazy - questions - high | 26 | 3919_lots_couple_crazy_questions |
| 3920 | supply - total - weeks - issues - benefits | 26 | 3920_supply_total_weeks_issues |
| 3921 | biggest - ecosystem - order - launch - swap | 26 | 3921_biggest_ecosystem_order_launch |
| 3922 | digital - right - trading - bitcoin - related | 26 | 3922_digital_right_trading_bitcoin |
| 3923 | guy - games - use - old - thing | 26 | 3923_guy_games_use_old |
| 3924 | event - week - trade - utc - trading | 26 | 3924_event_week_trade_utc |
| 3925 | short - term - long - higher - trading | 26 | 3925_short_term_long_higher |
| 3926 | upcoming - sign - information - trade - week | 26 | 3926_upcoming_sign_information_trade |
| 3927 | life - - - - | 26 | 3927_life___ |
| 3928 | chance - questions - lots - public - space | 26 | 3928_chance_questions_lots_public |
| 3929 | people - interesting - research - details - minimum | 26 | 3929_people_interesting_research_details |
| 3930 | comment - app - reward - daily - process | 26 | 3930_comment_app_reward_daily |
| 3931 | normal - block - fast - - | 26 | 3931_normal_block_fast_ |
| 3932 | today - good - people - money - job | 26 | 3932_today_good_people_money |
| 3933 | event - luck - sign - happy - thanks | 26 | 3933_event_luck_sign_happy |
| 3934 | option - site - secure - safe - friends | 26 | 3934_option_site_secure_safe |
| 3935 | latest - update - binance - bitcoin - transactions | 26 | 3935_latest_update_binance_bitcoin |
| 3936 | brother - lots - sir - level - guys | 26 | 3936_brother_lots_sir_level |
| 3937 | order - special - available - site - payment | 26 | 3937_order_special_available_site |
| 3938 | research - solution - blockchain - worth - protocol | 26 | 3938_research_solution_blockchain_worth |
| 3939 | brother - bit - help - dear - private | 26 | 3939_brother_bit_help_dear |
| 3940 | reason - account - user - earlier - reward | 26 | 3940_reason_account_user_earlier |
| 3941 | ai - bot - platform - feature - play | 26 | 3941_ai_bot_platform_feature |
| 3942 | alright - article - rate - better - late | 26 | 3942_alright_article_rate_better |
| 3943 | investing - glad - technology - right - attention | 26 | 3943_investing_glad_technology_right |
| 3944 | coins - active - online - hour - app | 26 | 3944_coins_active_online_hour |
| 3945 | glad - couple - family - job - option | 26 | 3945_glad_couple_family_job |
| 3946 | hour - quick - profit - contact - huge | 26 | 3946_hour_quick_profit_contact |
| 3947 | problems - past - alot - little - fake | 26 | 3947_problems_past_alot_little |
| 3948 | simple - special - yesterday - guy - issues | 26 | 3948_simple_special_yesterday_guy |
| 3949 | friends - level - friend - media - minimum | 26 | 3949_friends_level_friend_media |
| 3950 | dude - brother - close - mean - wan | 26 | 3950_dude_brother_close_mean |
| 3951 | period - plan - funds - - | 26 | 3951_period_plan_funds_ |
| 3952 | development - project - point - huge - plan | 26 | 3952_development_project_point_huge |
| 3953 | deal - type - team - way - buy | 26 | 3953_deal_type_team_way |
| 3954 | pool - holders - real - half - fees | 26 | 3954_pool_holders_real_half |
| 3955 | future - stay - tomorrow - note - rate | 26 | 3955_future_stay_tomorrow_note |
| 3956 | latest - - - - | 26 | 3956_latest___ |
| 3957 | experience - days - trading - connect - turn | 26 | 3957_experience_days_trading_connect |
| 3958 | message - - - - | 26 | 3958_message___ |
| 3959 | reply - comment - number - sir - contract | 26 | 3959_reply_comment_number_sir |
| 3960 | ta - near - major - listing - coins | 26 | 3960_ta_near_major_listing |
| 3961 | normal - protocol - order - set - coin | 26 | 3961_normal_protocol_order_set |
| 3962 | normal - difference - fair - moment - bad | 26 | 3962_normal_difference_fair_moment |
| 3963 | safe - - - - | 26 | 3963_safe___ |
| 3964 | reason - correct - power - - | 26 | 3964_reason_correct_power_ |
| 3965 | digital - attention - mind - end - safe | 26 | 3965_digital_attention_mind_end |
| 3966 | search - telegram - - - | 26 | 3966_search_telegram__ |
| 3967 | good - - - - | 26 | 3967_good___ |
| 3968 | haha - read - article - interested - correct | 26 | 3968_haha_read_article_interested |
| 3969 | total - reward - supply - liquidity - friends | 26 | 3969_total_reward_supply_liquidity |
| 3970 | end - love - project - good - great | 26 | 3970_end_love_project_good |
| 3971 | interesting - smart - point - comment - product | 26 | 3971_interesting_smart_point_comment |
| 3972 | major - benefits - fees - low - pool | 26 | 3972_major_benefits_fees_low |
| 3973 | - - - - | 26 | 3973____ |
| 3974 | trust - group - help - dm - support | 26 | 3974_trust_group_help_dm |
| 3975 | phone - issues - normal - problem - moment | 26 | 3975_phone_issues_normal_problem |
| 3976 | service - community - wallet - solution - future | 26 | 3976_service_community_wallet_solution |
| 3977 | business - app - supply - trading - money | 26 | 3977_business_app_supply_trading |
| 3978 | transaction - fees - recent - transactions - network | 26 | 3978_transaction_fees_recent_transactions |
| 3979 | article - digital - space - biggest - cryptocurrency | 26 | 3979_article_digital_space_biggest |
| 3980 | man - value - success - opportunity - chat | 26 | 3980_man_value_success_opportunity |
| 3981 | morning - dear - nice - transaction - work | 26 | 3981_morning_dear_nice_transaction |
| 3982 | certain - cause - single - members - case | 26 | 3982_certain_cause_single_members |
| 3983 | cool - okay - fake - nice - fun | 26 | 3983_cool_okay_fake_nice |
| 3984 | half - app - updates - crazy - weeks | 26 | 3984_half_app_updates_crazy |
| 3985 | market - price - - - | 26 | 3985_market_price__ |
| 3986 | word - payment - days - stop - new | 26 | 3986_word_payment_days_stop |
| 3987 | accounts - person - telegram - price - best | 26 | 3987_accounts_person_telegram_price |
| 3988 | protocol - situation - assets - current - funds | 25 | 3988_protocol_situation_assets_current |
| 3989 | chat - private - night - fun - thanks | 25 | 3989_chat_private_night_fun |
| 3990 | fair - main - situation - mate - ones | 25 | 3990_fair_main_situation_mate |
| 3991 | normal - chat - type - play - fine | 25 | 3991_normal_chat_type_play |
| 3992 | data - similar - correct - end - hard | 25 | 3992_data_similar_correct_end |
| 3993 | services - success - platform - right - investment | 25 | 3993_services_success_platform_right |
| 3994 | phone - number - code - country - app | 25 | 3994_phone_number_code_country |
| 3995 | bro - list - early - spot - massive | 25 | 3995_bro_list_early_spot |
| 3996 | interested - message - huge - page - post | 25 | 3996_interested_message_huge_page |
| 3997 | option - sir - profit - post - life | 25 | 3997_option_sir_profit_post |
| 3998 | search - telegram - - - | 25 | 3998_search_telegram__ |
| 3999 | year - different - release - tokens - upcoming | 25 | 3999_year_different_release_tokens |
| 4000 | liquidity - launch - tokens - contract - assets | 25 | 4000_liquidity_launch_tokens_contract |
| 4001 | country - digital - exchange - action - play | 25 | 4001_country_digital_exchange_action |
| 4002 | main - dude - specific - example - ways | 25 | 4002_main_dude_specific_example |
| 4003 | latest - morning - update - updates - smart | 25 | 4003_latest_morning_update_updates |
| 4004 | members - discord - accounts - ones - online | 25 | 4004_members_discord_accounts_ones |
| 4005 | talk - exchange - guys - - | 25 | 4005_talk_exchange_guys_ |
| 4006 | clear - bad - better - withdrawal - good | 25 | 4006_clear_bad_better_withdrawal |
| 4007 | - - - - | 25 | 4007____ |
| 4008 | daily - minimum - fee - reward - deposit | 25 | 4008_daily_minimum_fee_reward |
| 4009 | cause - attention - example - word - early | 25 | 4009_cause_attention_example_word |
| 4010 | deposit - spot - trading - trade - sir | 25 | 4010_deposit_spot_trading_trade |
| 4011 | love - life - work - people - step | 25 | 4011_love_life_work_people |
| 4012 | fun - dude - cool - years - true | 25 | 4012_fun_dude_cool_years |
| 4013 | word - point - private - service - works | 25 | 4013_word_point_private_service |
| 4014 | case - chart - daily - current - early | 25 | 4014_case_chart_daily_current |
| 4015 | idea - moment - sure - sense - head | 25 | 4015_idea_moment_sure_sense |
| 4016 | daily - video - reward - plans - higher | 25 | 4016_daily_video_reward_plans |
| 4017 | future - buying - near - feature - important | 25 | 4017_future_buying_near_feature |
| 4018 | wonder - special - ways - guy - home | 25 | 4018_wonder_special_ways_guy |
| 4019 | form - dm - funds - cool - work | 25 | 4019_form_dm_funds_cool |
| 4020 | members - group - available - message - telegram | 25 | 4020_members_group_available_message |
| 4021 | feature - correct - huge - current - years | 25 | 4021_feature_correct_huge_current |
| 4022 | big - life - group - - | 25 | 4022_big_life_group_ |
| 4023 | visit - head - special - normal - past | 25 | 4023_visit_head_special_normal |
| 4024 | - - - - | 25 | 4024____ |
| 4025 | services - easy - service - thanks - people | 25 | 4025_services_easy_service_thanks |
| 4026 | listing - tomorrow - pool - spot - deposit | 25 | 4026_listing_tomorrow_pool_spot |
| 4027 | links - multiple - main - official - real | 25 | 4027_links_multiple_main_official |
| 4028 | tho - attention - nice - job - months | 25 | 4028_tho_attention_nice_job |
| 4029 | night - - - - | 25 | 4029_night___ |
| 4030 | opinion - clear - main - bit - interested | 25 | 4030_opinion_clear_main_bit |
| 4031 | second - country - world - trade - success | 25 | 4031_second_country_world_trade |
| 4032 | technology - days - smart - media - admin | 25 | 4032_technology_days_smart_media |
| 4033 | payment - friend - company - real - happy | 25 | 4033_payment_friend_company_real |
| 4034 | huge - mate - sir - massive - ready | 25 | 4034_huge_mate_sir_massive |
| 4035 | rules - situation - global - ways - yesterday | 25 | 4035_rules_situation_global_ways |
| 4036 | guide - information - code - order - worth | 25 | 4036_guide_information_code_order |
| 4037 | better - dude - step - success - way | 25 | 4037_better_dude_step_success |
| 4038 | super - special - feature - mate - cause | 25 | 4038_super_special_feature_mate |
| 4039 | ama - question - details - wallet - welcome | 25 | 4039_ama_question_details_wallet |
| 4040 | media - social - month - page - trade | 25 | 4040_media_social_month_page |
| 4041 | today - bro - fine - morning - dear | 25 | 4041_today_bro_fine_morning |
| 4042 | - - - - | 25 | 4042____ |
| 4043 | app - visit - version - global - strong | 25 | 4043_app_visit_version_global |
| 4044 | word - example - correct - site - space | 25 | 4044_word_example_correct_site |
| 4045 | ppl - dude - nice - guy - bro | 25 | 4045_ppl_dude_nice_guy |
| 4046 | coins - coin - looks - later - guys | 25 | 4046_coins_coin_looks_later |
| 4047 | - - - - | 25 | 4047____ |
| 4048 | address - business - interested - available - sense | 25 | 4048_address_business_interested_available |
| 4049 | risk - reward - hand - high - rewards | 25 | 4049_risk_reward_hand_high |
| 4050 | code - latest - account - minimum - details | 25 | 4050_code_latest_account_minimum |
| 4051 | hand - late - fair - bad - fine | 25 | 4051_hand_late_fair_bad |
| 4052 | positive - bad - way - things - | 25 | 4052_positive_bad_way_things |
| 4053 | sense - strong - fair - massive - marketing | 25 | 4053_sense_strong_fair_massive |
| 4054 | yesterday - moment - likely - playing - guy | 25 | 4054_yesterday_moment_likely_playing |
| 4055 | control - hours - community - new - kind | 25 | 4055_control_hours_community_new |
| 4056 | global - assets - - - | 25 | 4056_global_assets__ |
| 4057 | major - set - solution - network - security | 25 | 4057_major_set_solution_network |
| 4058 | cryptocurrency - success - digital - country - global | 25 | 4058_cryptocurrency_success_digital_country |
| 4059 | fees - btc - high - fee - yesterday | 25 | 4059_fees_btc_high_fee |
| 4060 | fun - available - message - interested - dear | 25 | 4060_fun_available_message_interested |
| 4061 | market - specific - example - dev - issue | 25 | 4061_market_specific_example_dev |
| 4062 | global - fun - massive - launch - marketing | 25 | 4062_global_fun_massive_launch |
| 4063 | home - specific - special - exciting - reason | 25 | 4063_home_specific_special_exciting |
| 4064 | place - specific - article - control - stuff | 25 | 4064_place_specific_article_control |
| 4065 | swap - certain - event - reward - trade | 25 | 4065_swap_certain_event_reward |
| 4066 | late - moment - alot - course - development | 25 | 4066_late_moment_alot_course |
| 4067 | token - tokens - form - list - listing | 25 | 4067_token_tokens_form_list |
| 4068 | sense - article - drop - months - interesting | 25 | 4068_sense_article_drop_months |
| 4069 | users - business - people - support - | 25 | 4069_users_business_people_support |
| 4070 | list - dev - building - site - upcoming | 25 | 4070_list_dev_building_site |
| 4071 | live - platform - product - fact - web | 25 | 4071_live_platform_product_fact |
| 4072 | day - good - words - better - | 25 | 4072_day_good_words_better |
| 4073 | earn - invest - telegram - home - admin | 25 | 4073_earn_invest_telegram_home |
| 4074 | digital - problem - value - cryptocurrency - market | 25 | 4074_digital_problem_value_cryptocurrency |
| 4075 | night - - - - | 25 | 4075_night___ |
| 4076 | spot - example - worth - attention - idea | 25 | 4076_spot_example_worth_attention |
| 4077 | investing - successful - times - place - company | 25 | 4077_investing_successful_times_place |
| 4078 | stuff - channel - upcoming - night - idea | 25 | 4078_stuff_channel_upcoming_night |
| 4079 | media - social - limited - - | 25 | 4079_media_social_limited_ |
| 4080 | awesome - guy - rules - tomorrow - swap | 25 | 4080_awesome_guy_rules_tomorrow |
| 4081 | tho - code - product - write - simple | 25 | 4081_tho_code_product_write |
| 4082 | cryptocurrency - use - trading - crypto - time | 25 | 4082_cryptocurrency_use_trading_crypto |
| 4083 | features - friend - official - chain - fee | 25 | 4083_features_friend_official_chain |
| 4084 | protocol - latest - order - bot - news | 25 | 4084_protocol_latest_order_bot |
| 4085 | reply - fact - past - issues - fake | 25 | 4085_reply_fact_past_issues |
| 4086 | blockchain - projects - capital - latest - investors | 25 | 4086_blockchain_projects_capital_latest |
| 4087 | open - available - digital - account - pay | 25 | 4087_open_available_digital_account |
| 4088 | hours - new - days - reason - sir | 25 | 4088_hours_new_days_reason |
| 4089 | liquidity - recent - version - hand - protocol | 25 | 4089_liquidity_recent_version_hand |
| 4090 | night - fun - ready - chat - dude | 25 | 4090_night_fun_ready_chat |
| 4091 | comment - weeks - data - correct - plus | 25 | 4091_comment_weeks_data_correct |
| 4092 | year - btc - half - bitcoin - second | 25 | 4092_year_btc_half_bitcoin |
| 4093 | pool - app - likely - transaction - question | 25 | 4093_pool_app_likely_transaction |
| 4094 | bitcoin - assets - year - thought - mind | 25 | 4094_bitcoin_assets_year_thought |
| 4095 | guys - amazing - support - business - awesome | 25 | 4095_guys_amazing_support_business |
| 4096 | available - note - rate - payment - offer | 25 | 4096_available_note_rate_payment |
| 4097 | transaction - date - note - comment - details | 25 | 4097_transaction_date_note_comment |
| 4098 | total - number - report - increase - fees | 25 | 4098_total_number_report_increase |
| 4099 | fair - months - assets - digital - financial | 25 | 4099_fair_months_assets_digital |
| 4100 | projects - - - - | 25 | 4100_projects___ |
| 4101 | blockchain - ai - super - data - new | 25 | 4101_blockchain_ai_super_data |
| 4102 | ecosystem - welcome - future - chance - step | 25 | 4102_ecosystem_welcome_future_chance |
| 4103 | global - clear - app - rules - bitcoin | 25 | 4103_global_clear_app_rules |
| 4104 | multiple - - - - | 25 | 4104_multiple___ |
| 4105 | public - ai - bot - research - point | 25 | 4105_public_ai_bot_research |
| 4106 | market - bitcoin - crypto - key - total | 25 | 4106_market_bitcoin_crypto_key |
| 4107 | members - earlier - ones - kind - lot | 25 | 4107_members_earlier_ones_kind |
| 4108 | volume - assets - daily - - | 25 | 4108_volume_assets_daily_ |
| 4109 | early - job - message - awesome - user | 25 | 4109_early_job_message_awesome |
| 4110 | ai - blockchain - interesting - stuff - user | 25 | 4110_ai_blockchain_interesting_stuff |
| 4111 | simple - end - lol - higher - care | 25 | 4111_simple_end_lol_higher |
| 4112 | increase - higher - man - crazy - stuff | 25 | 4112_increase_higher_man_crazy |
| 4113 | recommend - run - ready - investment - | 25 | 4113_recommend_run_ready_investment |
| 4114 | interested - media - person - tomorrow - telegram | 25 | 4114_interested_media_person_tomorrow |
| 4115 | wallet - mate - stuff - order - friend | 25 | 4115_wallet_mate_stuff_order |
| 4116 | assets - digital - members - crypto - | 25 | 4116_assets_digital_members_crypto |
| 4117 | development - question - experience - projects - nft | 25 | 4117_development_question_experience_projects |
| 4118 | watch - list - certain - nice - ones | 25 | 4118_watch_list_certain_nice |
| 4119 | trade - digital - group - use - transactions | 25 | 4119_trade_digital_group_use |
| 4120 | message - days - - - | 25 | 4120_message_days__ |
| 4121 | funds - half - previous - alot - personal | 25 | 4121_funds_half_previous_alot |
| 4122 | tokens - new - - - | 25 | 4122_tokens_new__ |
| 4123 | recent - means - fees - blockchain - high | 24 | 4123_recent_means_fees_blockchain |
| 4124 | ama - binance - live - excited - special | 24 | 4124_ama_binance_live_excited |
| 4125 | community - tomorrow - weeks - kind - minutes | 24 | 4125_community_tomorrow_weeks_kind |
| 4126 | event - end - days - date - luck | 24 | 4126_event_end_days_date |
| 4127 | withdrawal - link - sign - members - group | 24 | 4127_withdrawal_link_sign_members |
| 4128 | late - join - big - - | 24 | 4128_late_join_big_ |
| 4129 | transaction - guide - web - sorry - look | 24 | 4129_transaction_guide_web_sorry |
| 4130 | data - public - network - blockchain - | 24 | 4130_data_public_network_blockchain |
| 4131 | face - words - fun - year - way | 24 | 4131_face_words_fun_year |
| 4132 | page - free - - - | 24 | 4132_page_free__ |
| 4133 | words - true - real - tbh - close | 24 | 4133_words_true_real_tbh |
| 4134 | swap - contract - binance - smart - support | 24 | 4134_swap_contract_binance_smart |
| 4135 | certain - members - right - support - today | 24 | 4135_certain_members_right_support |
| 4136 | home - family - true - na - course | 24 | 4136_home_family_true_na |
| 4137 | new - nice - early - love - members | 24 | 4137_new_nice_early_love |
| 4138 | plans - ready - network - binance - second | 24 | 4138_plans_ready_network_binance |
| 4139 | ecosystem - multiple - benefits - power - nft | 24 | 4139_ecosystem_multiple_benefits_power |
| 4140 | total - digital - assets - week - trade | 24 | 4140_total_digital_assets_week |
| 4141 | dm - old - real - guys - year | 24 | 4141_dm_old_real_guys |
| 4142 | technology - positive - protocol - ecosystem - year | 24 | 4142_technology_positive_protocol_ecosystem |
| 4143 | pls - eth - main - ethereum - site | 24 | 4143_pls_eth_main_ethereum |
| 4144 | win - risk - level - official - daily | 24 | 4144_win_risk_level_official |
| 4145 | fee - withdrawal - deposit - link - | 24 | 4145_fee_withdrawal_deposit_link |
| 4146 | protocol - control - past - morning - latest | 24 | 4146_protocol_control_past_morning |
| 4147 | binance - capital - early - course - offer | 24 | 4147_binance_capital_early_course |
| 4148 | information - free - message - today - new | 24 | 4148_information_free_message_today |
| 4149 | risk - level - official - daily - | 24 | 4149_risk_level_official_daily |
| 4150 | fine - build - problem - works - bit | 24 | 4150_fine_build_problem_works |
| 4151 | assets - report - trade - technology - crypto | 24 | 4151_assets_report_trade_technology |
| 4152 | dude - stuff - fine - words - love | 24 | 4152_dude_stuff_fine_words |
| 4153 | assets - users - months - years - announcement | 24 | 4153_assets_users_months_years |
| 4154 | care - dear - buddy - nice - day | 24 | 4154_care_dear_buddy_nice |
| 4155 | utc - period - app - official - website | 24 | 4155_utc_period_app_official |
| 4156 | reply - tho - thinking - type - person | 24 | 4156_reply_tho_thinking_type |
| 4157 | brother - man - rest - nice - little | 24 | 4157_brother_man_rest_nice |
| 4158 | answer - mate - ways - word - massive | 24 | 4158_answer_mate_ways_word |
| 4159 | exciting - holders - launch - huge - private | 24 | 4159_exciting_holders_launch_huge |
| 4160 | smart - - - - | 24 | 4160_smart___ |
| 4161 | fine - moment - man - matter - point | 24 | 4161_fine_moment_man_matter |
| 4162 | home - thing - month - people - | 24 | 4162_home_thing_month_people |
| 4163 | security - report - telegram - example - official | 24 | 4163_security_report_telegram_example |
| 4164 | ta - longer - opinion - biggest - tho | 24 | 4164_ta_longer_opinion_biggest |
| 4165 | believe - money - little - long - need | 24 | 4165_believe_money_little_long |
| 4166 | ethereum - guy - network - send - contract | 24 | 4166_ethereum_guy_network_send |
| 4167 | special - offer - business - correct - alot | 24 | 4167_special_offer_business_correct |
| 4168 | ones - financial - small - important - real | 24 | 4168_ones_financial_small_important |
| 4169 | face - investing - thanks - contact - sorry | 24 | 4169_face_investing_thanks_contact |
| 4170 | games - answer - drop - strong - small | 24 | 4170_games_answer_drop_strong |
| 4171 | users - big - crypto - space - recent | 24 | 4171_users_big_crypto_space |
| 4172 | night - interested - welcome - game - morning | 24 | 4172_night_interested_welcome_game |
| 4173 | man - time - process - - | 24 | 4173_man_time_process_ |
| 4174 | available - time - info - - | 24 | 4174_available_time_info_ |
| 4175 | past - year - coin - share - market | 24 | 4175_past_year_coin_share |
| 4176 | home - happy - money - thanks - week | 24 | 4176_home_happy_money_thanks |
| 4177 | late - complete - active - group - minutes | 24 | 4177_late_complete_active_group |
| 4178 | available - - - - | 24 | 4178_available___ |
| 4179 | sure - way - admin - dm - position | 24 | 4179_sure_way_admin_dm |
| 4180 | technology - dev - lots - rate - drop | 24 | 4180_technology_dev_lots_rate |
| 4181 | deal - dm - fun - person - number | 24 | 4181_deal_dm_fun_person |
| 4182 | cool - nice - point - reach - person | 24 | 4182_cool_nice_point_reach |
| 4183 | chart - family - thought - plan - marketing | 24 | 4183_chart_family_thought_plan |
| 4184 | search - network - positive - blockchain - family | 24 | 4184_search_network_positive_blockchain |
| 4185 | launch - new - platform - public - security | 24 | 4185_launch_new_platform_public |
| 4186 | cool - yesterday - idea - case - unique | 24 | 4186_cool_yesterday_idea_case |
| 4187 | drop - option - tho - luck - week | 24 | 4187_drop_option_tho_luck |
| 4188 | previous - idea - guide - current - latest | 24 | 4188_previous_idea_guide_current |
| 4189 | week - details - links - ama - complete | 24 | 4189_week_details_links_ama |
| 4190 | exciting - visit - cryptocurrency - news - chance | 24 | 4190_exciting_visit_cryptocurrency_news |
| 4191 | platform - - - - | 24 | 4191_platform___ |
| 4192 | certain - cause - kind - bit - different | 24 | 4192_certain_cause_kind_bit |
| 4193 | recent - person - data - cryptocurrency - news | 24 | 4193_recent_person_data_cryptocurrency |
| 4194 | nice - group - day - members - great | 24 | 4194_nice_group_day_members |
| 4195 | digital - likely - specific - exciting - potential | 24 | 4195_digital_likely_specific_exciting |
| 4196 | data - network - technology - specific - latest | 24 | 4196_data_network_technology_specific |
| 4197 | rules - questions - services - pm - group | 24 | 4197_rules_questions_services_pm |
| 4198 | profits - participate - trade - win - fun | 24 | 4198_profits_participate_trade_win |
| 4199 | morning - nice - good - friend - day | 24 | 4199_morning_nice_good_friend |
| 4200 | situation - current - term - mind - important | 24 | 4200_situation_current_term_mind |
| 4201 | fees - reward - transaction - btc - similar | 24 | 4201_fees_reward_transaction_btc |
| 4202 | matter - difference - crazy - upcoming - dev | 24 | 4202_matter_difference_crazy_upcoming |
| 4203 | example - public - control - point - late | 24 | 4203_example_public_control_point |
| 4204 | fee - fees - personal - clear - yesterday | 24 | 4204_fee_fees_personal_clear |
| 4205 | situation - article - normal - price - weeks | 24 | 4205_situation_article_normal_price |
| 4206 | btc - trade - - - | 24 | 4206_btc_trade__ |
| 4207 | cryptocurrency - way - opinion - time - example | 24 | 4207_cryptocurrency_way_opinion_time |
| 4208 | huge - accounts - available - active - channel | 24 | 4208_huge_accounts_available_active |
| 4209 | dude - difference - correct - guy - type | 24 | 4209_dude_difference_correct_guy |
| 4210 | assets - normal - issue - problem - option | 24 | 4210_assets_normal_issue_problem |
| 4211 | latest - updates - message - active - huge | 24 | 4211_latest_updates_message_active |
| 4212 | - - - - | 24 | 4212____ |
| 4213 | - - - - | 24 | 4213____ |
| 4214 | assets - funds - digital - development - different | 24 | 4214_assets_funds_digital_development |
| 4215 | pool - reward - total - supply - article | 24 | 4215_pool_reward_total_supply |
| 4216 | drop - action - potential - price - profits | 24 | 4216_drop_action_potential_price |
| 4217 | sir - step - life - family - glad | 24 | 4217_sir_step_life_family |
| 4218 | transactions - previous - solution - number - transaction | 24 | 4218_transactions_previous_solution_number |
| 4219 | key - bitcoin - need - - | 24 | 4219_key_bitcoin_need_ |
| 4220 | correct - sorry - little - opportunity - good | 24 | 4220_correct_sorry_little_opportunity |
| 4221 | fair - deal - account - post - personal | 24 | 4221_fair_deal_account_post |
| 4222 | pm - ready - wallet - money - best | 24 | 4222_pm_ready_wallet_money |
| 4223 | btc - price - bitcoin - support - level | 24 | 4223_btc_price_bitcoin_support |
| 4224 | - - - - | 24 | 4224____ |
| 4225 | words - man - happy - thanks - help | 24 | 4225_words_man_happy_thanks |
| 4226 | way - right - - - | 24 | 4226_way_right__ |
| 4227 | excited - profits - payment - work - wallet | 24 | 4227_excited_profits_payment_work |
| 4228 | super - special - win - late - number | 24 | 4228_super_special_win_late |
| 4229 | version - current - launch - official - code | 24 | 4229_version_current_launch_official |
| 4230 | latest - updates - links - news - official | 24 | 4230_latest_updates_links_news |
| 4231 | risk - projects - - - | 24 | 4231_risk_projects__ |
| 4232 | home - report - funds - global - private | 24 | 4232_home_report_funds_global |
| 4233 | words - contact - safe - happy - lot | 24 | 4233_words_contact_safe_happy |
| 4234 | open - couple - example - ones - code | 24 | 4234_open_couple_example_ones |
| 4235 | glad - man - sir - team - wrong | 24 | 4235_glad_man_sir_team |
| 4236 | position - latest - little - members - welcome | 24 | 4236_position_latest_little_members |
| 4237 | blockchain - - - - | 24 | 4237_blockchain___ |
| 4238 | digital - features - plans - - | 24 | 4238_digital_features_plans_ |
| 4239 | dude - excited - bit - course - year | 24 | 4239_dude_excited_bit_course |
| 4240 | certain - feature - run - situation - control | 24 | 4240_certain_feature_run_situation |
| 4241 | dm - drop - - - | 24 | 4241_dm_drop__ |
| 4242 | home - interested - current - online - links | 24 | 4242_home_interested_current_online |
| 4243 | ai - blockchain - future - opinion - ones | 24 | 4243_ai_blockchain_future_opinion |
| 4244 | couple - running - pool - luck - little | 24 | 4244_couple_running_pool_luck |
| 4245 | type - protocol - solution - stuff - chance | 24 | 4245_type_protocol_solution_stuff |
| 4246 | financial - service - biggest - kind - services | 24 | 4246_financial_service_biggest_kind |
| 4247 | pool - rewards - issue - attention - reward | 24 | 4247_pool_rewards_issue_attention |
| 4248 | bit - example - luck - minutes - case | 24 | 4248_bit_example_luck_minutes |
| 4249 | security - projects - project - important - kind | 24 | 4249_security_projects_project_important |
| 4250 | trust - ai - blockchain - problems - development | 24 | 4250_trust_ai_blockchain_problems |
| 4251 | ai - guy - smart - case - right | 24 | 4251_ai_guy_smart_case |
| 4252 | event - ones - hour - end - month | 24 | 4252_event_ones_hour_end |
| 4253 | solution - idea - data - nice - making | 24 | 4253_solution_idea_data_nice |
| 4254 | bot - question - answer - research - telegram | 24 | 4254_bot_question_answer_research |
| 4255 | way - ways - best - bit - course | 24 | 4255_way_ways_best_bit |
| 4256 | word - half - near - tho - option | 24 | 4256_word_half_near_tho |
| 4257 | clear - focus - point - dear - person | 24 | 4257_clear_focus_point_dear |
| 4258 | haha - correct - bad - key - mate | 24 | 4258_haha_correct_bad_key |
| 4259 | word - telegram - channel - payment - worth | 24 | 4259_word_telegram_channel_payment |
| 4260 | buddy - mate - great - sir - day | 24 | 4260_buddy_mate_great_sir |
| 4261 | friend - channel - info - free - link | 24 | 4261_friend_channel_info_free |
| 4262 | news - - - - | 24 | 4262_news___ |
| 4263 | liquidity - sense - normal - example - fair | 24 | 4263_liquidity_sense_normal_example |
| 4264 | - - - - | 24 | 4264____ |
| 4265 | moment - btc - news - example - bitcoin | 24 | 4265_moment_btc_news_example |
| 4266 | works - features - updates - option - accounts | 24 | 4266_works_features_updates_option |
| 4267 | time - good - past - dm - minutes | 24 | 4267_time_good_past_dm |
| 4268 | financial - technology - experience - development - product | 24 | 4268_financial_technology_experience_development |
| 4269 | assets - close - cool - positive - idea | 24 | 4269_assets_close_cool_positive |
| 4270 | media - social - limited - - | 24 | 4270_media_social_limited_ |
| 4271 | games - good - time - future - game | 24 | 4271_games_good_time_future |
| 4272 | watch - list - - - | 24 | 4272_watch_list__ |
| 4273 | questions - funds - members - information - chat | 24 | 4273_questions_funds_members_information |
| 4274 | excited - home - god - care - family | 24 | 4274_excited_home_god_care |
| 4275 | face - question - experience - lot - available | 24 | 4275_face_question_experience_lot |
| 4276 | - - - - | 24 | 4276____ |
| 4277 | upcoming - important - increase - product - development | 24 | 4277_upcoming_important_increase_product |
| 4278 | close - fair - option - enter - home | 24 | 4278_close_fair_option_enter |
| 4279 | bitcoin - team - new - - | 24 | 4279_bitcoin_team_new_ |
| 4280 | tokens - small - blockchain - fee - value | 24 | 4280_tokens_small_blockchain_fee |
| 4281 | features - research - users - support - web | 24 | 4281_features_research_users_support |
| 4282 | release - report - open - ai - research | 24 | 4282_release_report_open_ai |
| 4283 | marketing - listing - launch - huge - contract | 24 | 4283_marketing_listing_launch_huge |
| 4284 | fees - fee - low - guys - chain | 24 | 4284_fees_fee_low_guys |
| 4285 | friend - dude - friends - problems - ur | 24 | 4285_friend_dude_friends_problems |
| 4286 | previous - launch - projects - marketing - low | 24 | 4286_previous_launch_projects_marketing |
| 4287 | major - past - global - data - market | 24 | 4287_major_past_global_data |
| 4288 | social - different - major - high - guys | 24 | 4288_social_different_major_high |
| 4289 | country - cool - fine - real - world | 24 | 4289_country_cool_fine_real |
| 4290 | thanks - sorry - website - free - | 24 | 4290_thanks_sorry_website_free |
| 4291 | crazy - awesome - love - shit - place | 24 | 4291_crazy_awesome_love_shit |
| 4292 | post - normal - links - recent - list | 24 | 4292_post_normal_links_recent |
| 4293 | development - contract - liquidity - web - launch | 24 | 4293_development_contract_liquidity_web |
| 4294 | period - app - official - website - deposit | 24 | 4294_period_app_official_website |
| 4295 | private - financial - security - end - chain | 24 | 4295_private_financial_security_end |
| 4296 | single - man - problems - problem - issues | 24 | 4296_single_man_problems_problem |
| 4297 | supply - address - option - words - bit | 24 | 4297_supply_address_option_words |
| 4298 | sure - times - little - issue - level | 24 | 4298_sure_times_little_issue |
| 4299 | home - rest - turn - word - night | 24 | 4299_home_rest_turn_word |
| 4300 | brother - head - listing - crazy - bit | 24 | 4300_brother_head_listing_crazy |
| 4301 | video - solution - option - main - home | 24 | 4301_video_solution_option_main |
| 4302 | open - years - link - - | 24 | 4302_open_years_link_ |
| 4303 | projects - future - new - coin - | 24 | 4303_projects_future_new_coin |
| 4304 | case - friend - way - look - | 24 | 4304_case_friend_way_look |
| 4305 | current - future - price - available - buy | 24 | 4305_current_future_price_available |
| 4306 | bitcoin - interested - ways - lots - current | 23 | 4306_bitcoin_interested_ways_lots |
| 4307 | btc - - - - | 23 | 4307_btc___ |
| 4308 | hour - transactions - volume - liquidity - price | 23 | 4308_hour_transactions_volume_liquidity |
| 4309 | words - contact - safe - happy - lot | 23 | 4309_words_contact_safe_happy |
| 4310 | opinion - people - lot - crazy - main | 23 | 4310_opinion_people_lot_crazy |
| 4311 | bit - difference - fact - guy - ones | 23 | 4311_bit_difference_fact_guy |
| 4312 | gon - option - minutes - game - issue | 23 | 4312_gon_option_minutes_game |
| 4313 | bitcoin - second - - - | 23 | 4313_bitcoin_second__ |
| 4314 | likely - situation - possible - type - mate | 23 | 4314_likely_situation_possible_type |
| 4315 | example - number - order - drop - updates | 23 | 4315_example_number_order_drop |
| 4316 | daily - members - official - friends - group | 23 | 4316_daily_members_official_friends |
| 4317 | available - today - - - | 23 | 4317_available_today__ |
| 4318 | utc - link - version - - | 23 | 4318_utc_link_version_ |
| 4319 | ama - period - rate - reward - risk | 23 | 4319_ama_period_rate_reward |
| 4320 | control - ethereum - bitcoin - - | 23 | 4320_control_ethereum_bitcoin_ |
| 4321 | country - months - - - | 23 | 4321_country_months__ |
| 4322 | dev - thing - people - users - | 23 | 4322_dev_thing_people_users |
| 4323 | feature - volume - safe - app - high | 23 | 4323_feature_volume_safe_app |
| 4324 | point - - - - | 23 | 4324_point___ |
| 4325 | price - bitcoin - new - yesterday - tomorrow | 23 | 4325_price_bitcoin_new_yesterday |
| 4326 | price - end - year - break - moment | 23 | 4326_price_end_year_break |
| 4327 | dude - brother - crazy - alot - cool | 23 | 4327_dude_brother_crazy_alot |
| 4328 | contact - fake - open - deal - members | 23 | 4328_contact_fake_open_deal |
| 4329 | key - private - luck - swap - hard | 23 | 4329_key_private_luck_swap |
| 4330 | head - control - mind - interesting - research | 23 | 4330_head_control_mind_interesting |
| 4331 | country - special - day - morning - happy | 23 | 4331_country_special_day_morning |
| 4332 | version - exchange - network - ecosystem - volume | 23 | 4332_version_exchange_network_ecosystem |
| 4333 | great - friends - guys - good - day | 23 | 4333_great_friends_guys_good |
| 4334 | block - question - future - community - | 23 | 4334_block_question_future_community |
| 4335 | rewards - working - week - days - specific | 23 | 4335_rewards_working_week_days |
| 4336 | today - fine - amazing - day - team | 23 | 4336_today_fine_amazing_day |
| 4337 | technology - unique - user - network - excited | 23 | 4337_technology_unique_user_network |
| 4338 | code - updates - app - fees - coins | 23 | 4338_code_updates_app_fees |
| 4339 | spot - latest - sign - services - profits | 23 | 4339_spot_latest_sign_services |
| 4340 | pool - claim - extra - trading - new | 23 | 4340_pool_claim_extra_trading |
| 4341 | old - couple - chat - real - single | 23 | 4341_old_couple_chat_real |
| 4342 | nice - wish - true - huge - fake | 23 | 4342_nice_wish_true_huge |
| 4343 | kind - person - man - - | 23 | 4343_kind_person_man_ |
| 4344 | specific - utc - announcement - listing - coin | 23 | 4344_specific_utc_announcement_listing |
| 4345 | payment - recent - focus - global - version | 23 | 4345_payment_recent_focus_global |
| 4346 | users - funds - news - - | 23 | 4346_users_funds_news_ |
| 4347 | quick - answer - reply - thanks - sir | 23 | 4347_quick_answer_reply_thanks |
| 4348 | couple - details - latest - interested - person | 23 | 4348_couple_details_latest_interested |
| 4349 | users - ecosystem - words - coin - coins | 23 | 4349_users_ecosystem_words_coin |
| 4350 | biggest - date - world - way - money | 23 | 4350_biggest_date_world_way |
| 4351 | strong - moment - situation - buddy - period | 23 | 4351_strong_moment_situation_buddy |
| 4352 | morning - hope - good - brother - guys | 23 | 4352_morning_hope_good_brother |
| 4353 | pm - - - - | 23 | 4353_pm___ |
| 4354 | yesterday - country - hour - mate - important | 23 | 4354_yesterday_country_hour_mate |
| 4355 | similar - assets - solution - access - product | 23 | 4355_similar_assets_solution_access |
| 4356 | higher - person - capital - sure - hours | 23 | 4356_higher_person_capital_sure |
| 4357 | term - rate - short - legit - plans | 23 | 4357_term_rate_short_legit |
| 4358 | risk - win - level - official - daily | 23 | 4358_risk_win_level_official |
| 4359 | past - year - value - example - massive | 23 | 4359_past_year_value_example |
| 4360 | ecosystem - chain - things - week - liquidity | 23 | 4360_ecosystem_chain_things_week |
| 4361 | web - holders - liquidity - set - risk | 23 | 4361_web_holders_liquidity_set |
| 4362 | ways - project - projects - important - complete | 23 | 4362_ways_project_projects_important |
| 4363 | discord - main - official - telegram - world | 23 | 4363_discord_main_official_telegram |
| 4364 | step - reward - complete - send - register | 23 | 4364_step_reward_complete_send |
| 4365 | difference - second - wrong - person - bro | 23 | 4365_difference_second_wrong_person |
| 4366 | market - good - today - interested - open | 23 | 4366_market_good_today_interested |
| 4367 | - - - - | 23 | 4367____ |
| 4368 | risk - level - official - daily - | 23 | 4368_risk_level_official_daily |
| 4369 | binance - coins - coin - old - dev | 23 | 4369_binance_coins_coin_old |
| 4370 | kind - person - main - personal - support | 23 | 4370_kind_person_main_personal |
| 4371 | ones - search - answer - number - plans | 23 | 4371_ones_search_answer_number |
| 4372 | product - similar - awesome - kind - chart | 23 | 4372_product_similar_awesome_kind |
| 4373 | chat - admin - personal - second - sorry | 23 | 4373_chat_admin_personal_second |
| 4374 | buddy - home - friend - bro - mate | 23 | 4374_buddy_home_friend_bro |
| 4375 | night - - - - | 23 | 4375_night___ |
| 4376 | bot - rules - pool - luck - share | 23 | 4376_bot_rules_pool_luck |
| 4377 | buying - brother - bit - lol - sir | 23 | 4377_buying_brother_bit_lol |
| 4378 | - - - - | 23 | 4378____ |
| 4379 | guy - fine - second - big - end | 23 | 4379_guy_fine_second_big |
| 4380 | hour - minimum - launch - payment - crypto | 23 | 4380_hour_minimum_launch_payment |
| 4381 | idea - cool - course - ecosystem - type | 23 | 4381_idea_cool_course_ecosystem |
| 4382 | bit - glad - post - link - | 23 | 4382_bit_glad_post_link |
| 4383 | clear - digital - trading - investors - market | 23 | 4383_clear_digital_trading_investors |
| 4384 | key - visit - investment - profit - option | 23 | 4384_key_visit_investment_profit |
| 4385 | worth - financial - report - latest - months | 23 | 4385_worth_financial_report_latest |
| 4386 | sense - type - fine - second - thing | 23 | 4386_sense_type_fine_second |
| 4387 | thanks - dude - bro - mate - kind | 23 | 4387_thanks_dude_bro_mate |
| 4388 | pm - minimum - fine - daily - details | 23 | 4388_pm_minimum_fine_daily |
| 4389 | latest - account - community - free - link | 23 | 4389_latest_account_community_free |
| 4390 | biggest - term - chance - problem - plans | 23 | 4390_biggest_term_chance_problem |
| 4391 | version - high - correct - moment - end | 23 | 4391_version_high_correct_moment |
| 4392 | home - plan - process - game - unique | 23 | 4392_home_plan_process_game |
| 4393 | version - app - global - times - different | 23 | 4393_version_app_global_times |
| 4394 | potential - access - crypto - group - | 23 | 4394_potential_access_crypto_group |
| 4395 | members - list - active - point - fact | 23 | 4395_members_list_active_point |
| 4396 | money - normal - friend - dear - way | 23 | 4396_money_normal_friend_dear |
| 4397 | recent - ecosystem - article - love - projects | 23 | 4397_recent_ecosystem_article_love |
| 4398 | half - sense - shit - man - minutes | 23 | 4398_half_sense_shit_man |
| 4399 | bot - available - - - | 23 | 4399_bot_available__ |
| 4400 | personal - accounts - key - private - information | 23 | 4400_personal_accounts_key_private |
| 4401 | minimum - total - deposit - withdraw - days | 23 | 4401_minimum_total_deposit_withdraw |
| 4402 | left - visit - order - app - live | 23 | 4402_left_visit_order_app |
| 4403 | nice - things - tho - guy - fun | 23 | 4403_nice_things_tho_guy |
| 4404 | fun - interested - old - year - love | 23 | 4404_fun_interested_old_year |
| 4405 | case - news - months - team - fees | 23 | 4405_case_news_months_team |
| 4406 | plans - today - plan - mate - day | 23 | 4406_plans_today_plan_mate |
| 4407 | ethereum - fee - safe - - | 23 | 4407_ethereum_fee_safe_ |
| 4408 | channel - telegram - hours - real - start | 23 | 4408_channel_telegram_hours_real |
| 4409 | blockchain - ecosystem - kind - defi - awesome | 23 | 4409_blockchain_ecosystem_kind_defi |
| 4410 | position - kinda - similar - personal - bit | 23 | 4410_position_kinda_similar_personal |
| 4411 | fact - number - ones - early - bit | 23 | 4411_fact_number_ones_early |
| 4412 | - - - - | 23 | 4412____ |
| 4413 | example - similar - spot - opinion - point | 23 | 4413_example_similar_spot_opinion |
| 4414 | normal - small - low - true - bad | 23 | 4414_normal_small_low_true |
| 4415 | upcoming - fair - attention - token - listing | 23 | 4415_upcoming_fair_attention_token |
| 4416 | features - latest - version - turn - word | 23 | 4416_features_latest_version_turn |
| 4417 | wish - lol - chance - little - man | 23 | 4417_wish_lol_chance_little |
| 4418 | life - best - point - chance - good | 23 | 4418_life_best_point_chance |
| 4419 | protocol - discord - app - article - information | 23 | 4419_protocol_discord_app_article |
| 4420 | glad - ecosystem - news - web - event | 23 | 4420_glad_ecosystem_news_web |
| 4421 | tho - friend - bad - things - nice | 23 | 4421_tho_friend_bad_things |
| 4422 | year - launch - huge - low - potential | 23 | 4422_year_launch_huge_low |
| 4423 | current - price - - - | 23 | 4423_current_price__ |
| 4424 | reward - minimum - friends - market - price | 23 | 4424_reward_minimum_friends_market |
| 4425 | hard - situation - biggest - shit - guys | 23 | 4425_hard_situation_biggest_shit |
| 4426 | true - rest - single - kind - answer | 23 | 4426_true_rest_single_kind |
| 4427 | crypto - half - previous - close - protocol | 23 | 4427_crypto_half_previous_close |
| 4428 | break - weeks - guy - total - building | 23 | 4428_break_weeks_guy_total |
| 4429 | price - couple - increase - ones - fine | 23 | 4429_price_couple_increase_ones |
| 4430 | pepe - unique - fair - game - chance | 23 | 4430_pepe_unique_fair_game |
| 4431 | game - tomorrow - thinking - yesterday - true | 23 | 4431_game_tomorrow_thinking_yesterday |
| 4432 | half - super - technology - sense - following | 23 | 4432_half_super_technology_sense |
| 4433 | accounts - fees - coin - investors - bitcoin | 23 | 4433_accounts_fees_coin_investors |
| 4434 | hand - lol - normal - term - person | 23 | 4434_hand_lol_normal_term |
| 4435 | private - telegram - massive - block - cool | 23 | 4435_private_telegram_massive_block |
| 4436 | transactions - coins - tokens - wallet - network | 23 | 4436_transactions_coins_tokens_wallet |
| 4437 | day - - - - | 23 | 4437_day___ |
| 4438 | hour - half - couple - hours - weeks | 23 | 4438_hour_half_couple_hours |
| 4439 | good - awesome - mate - thing - best | 23 | 4439_good_awesome_mate_thing |
| 4440 | free - crazy - interesting - social - channel | 23 | 4440_free_crazy_interesting_social |
| 4441 | case - number - related - reply - mind | 23 | 4441_case_number_related_reply |
| 4442 | bitcoin - point - price - - | 23 | 4442_bitcoin_point_price_ |
| 4443 | excited - type - cryptocurrency - fee - extra | 23 | 4443_excited_type_cryptocurrency_fee |
| 4444 | recent - visit - tokens - add - network | 23 | 4444_recent_visit_tokens_add |
| 4445 | buying - sense - showing - moment - massive | 23 | 4445_buying_sense_showing_moment |
| 4446 | main - couple - stuff - period - protocol | 23 | 4446_main_couple_stuff_period |
| 4447 | telegram - feature - option - important - app | 23 | 4447_telegram_feature_option_important |
| 4448 | case - crypto - certain - difference - research | 23 | 4448_case_crypto_certain_difference |
| 4449 | position - volume - market - price - trading | 23 | 4449_position_volume_market_price |
| 4450 | fact - good - attention - main - things | 23 | 4450_fact_good_attention_main |
| 4451 | online - work - country - home - job | 23 | 4451_online_work_country_home |
| 4452 | past - focus - deal - benefits - try | 23 | 4452_past_focus_deal_benefits |
| 4453 | technology - projects - capital - funds - | 23 | 4453_technology_projects_capital_funds |
| 4454 | ur - thing - dear - reason - admin | 23 | 4454_ur_thing_dear_reason |
| 4455 | listing - trading - new - yesterday - users | 23 | 4455_listing_trading_new_yesterday |
| 4456 | case - phone - issues - deal - okay | 23 | 4456_case_phone_issues_deal |
| 4457 | happy - excited - free - time - | 23 | 4457_happy_excited_free_time |
| 4458 | ecosystem - defi - key - swap - rewards | 23 | 4458_ecosystem_defi_key_swap |
| 4459 | guy - situation - release - period - bit | 23 | 4459_guy_situation_release_period |
| 4460 | course - claim - tokens - event - technology | 23 | 4460_course_claim_tokens_event |
| 4461 | btc - number - - - | 23 | 4461_btc_number__ |
| 4462 | guess - hand - bit - answer - guy | 23 | 4462_guess_hand_bit_answer |
| 4463 | bitcoin - biggest - interesting - single - issue | 23 | 4463_bitcoin_biggest_interesting_single |
| 4464 | deal - research - action - recent - latest | 23 | 4464_deal_research_action_recent |
| 4465 | exchange - answer - tokens - morning - moment | 23 | 4465_exchange_answer_tokens_morning |
| 4466 | thought - damn - gone - okay - yesterday | 23 | 4466_thought_damn_gone_okay |
| 4467 | stop - world - best - day - good | 23 | 4467_stop_world_best_day |
| 4468 | exchange - assets - crypto - cryptocurrency - early | 23 | 4468_exchange_assets_crypto_cryptocurrency |
| 4469 | article - contract - tokens - period - announcement | 23 | 4469_article_contract_tokens_period |
| 4470 | article - guys - work - - | 23 | 4470_article_guys_work_ |
| 4471 | - - - - | 23 | 4471____ |
| 4472 | dude - fix - game - awesome - issue | 23 | 4472_dude_fix_game_awesome |
| 4473 | good - - - - | 23 | 4473_good___ |
| 4474 | risk - level - official - daily - | 23 | 4474_risk_level_official_daily |
| 4475 | major - hand - certain - head - volume | 23 | 4475_major_hand_certain_head |
| 4476 | liquidity - potential - price - bitcoin - sense | 23 | 4476_liquidity_potential_price_bitcoin |
| 4477 | token - safe - low - launch - great | 23 | 4477_token_safe_low_launch |
| 4478 | users - - - - | 23 | 4478_users___ |
| 4479 | crypto - video - space - awesome - stuff | 23 | 4479_crypto_video_space_awesome |
| 4480 | country - second - platform - plans - end | 22 | 4480_country_second_platform_plans |
| 4481 | links - free - payment - multiple - low | 22 | 4481_links_free_payment_multiple |
| 4482 | smart - market - course - important - power | 22 | 4482_smart_market_course_important |
| 4483 | major - past - correct - post - account | 22 | 4483_major_past_correct_post |
| 4484 | questions - - - - | 22 | 4484_questions___ |
| 4485 | ecosystem - discord - glad - luck - live | 22 | 4485_ecosystem_discord_glad_luck |
| 4486 | action - article - okay - note - fake | 22 | 4486_action_article_okay_note |
| 4487 | personal - fun - questions - services - projects | 22 | 4487_personal_fun_questions_services |
| 4488 | stuff - awesome - crazy - pretty - night | 22 | 4488_stuff_awesome_crazy_pretty |
| 4489 | action - message - - - | 22 | 4489_action_message__ |
| 4490 | sign - bitcoin - certain - change - | 22 | 4490_sign_bitcoin_certain_change |
| 4491 | network - trust - feature - hit - ecosystem | 22 | 4491_network_trust_feature_hit |
| 4492 | updates - update - mate - latest - week | 22 | 4492_updates_update_mate_latest |
| 4493 | supply - total - chain - funds - near | 22 | 4493_supply_total_chain_funds |
| 4494 | product - search - small - fake - app | 22 | 4494_product_search_small_fake |
| 4495 | yesterday - minutes - current - update - week | 22 | 4495_yesterday_minutes_current_update |
| 4496 | action - head - user - capital - | 22 | 4496_action_head_user_capital |
| 4497 | night - fine - hour - tomorrow - amazing | 22 | 4497_night_fine_hour_tomorrow |
| 4498 | related - assets - features - close - note | 22 | 4498_related_assets_features_close |
| 4499 | links - private - message - members - group | 22 | 4499_links_private_message_members |
| 4500 | interested - - - - | 22 | 4500_interested___ |
| 4501 | tho - stuff - personal - course - dear | 22 | 4501_tho_stuff_personal_course |
| 4502 | minutes - bot - network - - | 22 | 4502_minutes_bot_network_ |
| 4503 | attention - exciting - place - opportunity - tokens | 22 | 4503_attention_exciting_place_opportunity |
| 4504 | total - token - guys - big - | 22 | 4504_total_token_guys_big |
| 4505 | massive - minutes - family - true - man | 22 | 4505_massive_minutes_family_true |
| 4506 | certain - pepe - users - face - write | 22 | 4506_certain_pepe_users_face |
| 4507 | plan - issues - order - capital - days | 22 | 4507_plan_issues_order_capital |
| 4508 | fun - little - ready - good - | 22 | 4508_fun_little_ready_good |
| 4509 | position - person - right - project - company | 22 | 4509_position_person_right_project |
| 4510 | features - ethereum - difference - network - private | 22 | 4510_features_ethereum_difference_network |
| 4511 | price - - - - | 22 | 4511_price___ |
| 4512 | game - profit - business - months - people | 22 | 4512_game_profit_business_months |
| 4513 | ecosystem - great - different - place - token | 22 | 4513_ecosystem_great_different_place |
| 4514 | claim - tokens - chain - community - active | 22 | 4514_claim_tokens_chain_community |
| 4515 | questions - discord - feel - information - channel | 22 | 4515_questions_discord_feel_information |
| 4516 | plans - community - opinion - wrong - | 22 | 4516_plans_community_opinion_wrong |
| 4517 | moment - ama - known - alot - reward | 22 | 4517_moment_ama_known_alot |
| 4518 | user - example - post - account - | 22 | 4518_user_example_post_account |
| 4519 | assets - projects - real - research - fun | 22 | 4519_assets_projects_real_research |
| 4520 | bit - guys - - - | 22 | 4520_bit_guys__ |
| 4521 | - - - - | 22 | 4521____ |
| 4522 | little - early - mind - bad - thing | 22 | 4522_little_early_mind_bad |
| 4523 | reason - price - bitcoin - - | 22 | 4523_reason_price_bitcoin_ |
| 4524 | - - - - | 22 | 4524____ |
| 4525 | luck - stuff - night - sorry - rate | 22 | 4525_luck_stuff_night_sorry |
| 4526 | fake - - - - | 22 | 4526_fake___ |
| 4527 | transaction - worth - years - bitcoin - crypto | 22 | 4527_transaction_worth_years_bitcoin |
| 4528 | focus - development - ecosystem - easy - users | 22 | 4528_focus_development_ecosystem_easy |
| 4529 | global - app - users - official - access | 22 | 4529_global_app_users_official |
| 4530 | exciting - latest - ready - games - data | 22 | 4530_exciting_latest_ready_games |
| 4531 | words - sure - date - man - works | 22 | 4531_words_sure_date_man |
| 4532 | capital - deal - dude - new - course | 22 | 4532_capital_deal_dude_new |
| 4533 | point - price - - - | 22 | 4533_point_price__ |
| 4534 | mate - exchange - buddy - ones - trust | 22 | 4534_mate_exchange_buddy_ones |
| 4535 | love - time - support - glad - rewards | 22 | 4535_love_time_support_glad |
| 4536 | success - order - point - hard - work | 22 | 4536_success_order_point_hard |
| 4537 | minimum - simple - daily - user - contact | 22 | 4537_minimum_simple_daily_user |
| 4538 | home - care - family - sir - step | 22 | 4538_home_care_family_sir |
| 4539 | news - bitcoin - - - | 22 | 4539_news_bitcoin__ |
| 4540 | welcome - members - questions - new - free | 22 | 4540_welcome_members_questions_new |
| 4541 | specific - interesting - technology - type - personal | 22 | 4541_specific_interesting_technology_type |
| 4542 | financial - control - solution - ones - power | 22 | 4542_financial_control_solution_ones |
| 4543 | bad - short - space - times - different | 22 | 4543_bad_short_space_times |
| 4544 | chart - discord - word - help - members | 22 | 4544_chart_discord_word_help |
| 4545 | assets - year - guy - months - worth | 22 | 4545_assets_year_guy_months |
| 4546 | community - links - link - interesting - man | 22 | 4546_community_links_link_interesting |
| 4547 | hand - country - fact - problem - wrong | 22 | 4547_hand_country_fact_problem |
| 4548 | idea - man - future - guys - great | 22 | 4548_idea_man_future_guys |
| 4549 | awesome - perfect - quick - secure - way | 22 | 4549_awesome_perfect_quick_secure |
| 4550 | media - page - social - available - | 22 | 4550_media_page_social_available |
| 4551 | claim - site - tokens - transactions - opportunity | 22 | 4551_claim_site_tokens_transactions |
| 4552 | game - luck - details - experience - new | 22 | 4552_game_luck_details_experience |
| 4553 | latest - bot - contract - launch - sure | 22 | 4553_latest_bot_contract_launch |
| 4554 | form - minimum - tomorrow - withdrawal - legit | 22 | 4554_form_minimum_tomorrow_withdrawal |
| 4555 | luck - opportunity - profit - option - exciting | 22 | 4555_luck_opportunity_profit_option |
| 4556 | excited - profit - admin - profits - rewards | 22 | 4556_excited_profit_admin_profits |
| 4557 | report - run - news - video - token | 22 | 4557_report_run_news_video |
| 4558 | night - details - - - | 22 | 4558_night_details__ |
| 4559 | mind - reason - different - platform - profit | 22 | 4559_mind_reason_different_platform |
| 4560 | simple - form - guide - ecosystem - luck | 22 | 4560_simple_form_guide_ecosystem |
| 4561 | search - type - information - unique - channel | 22 | 4561_search_type_information_unique |
| 4562 | private - members - team - situation - support | 22 | 4562_private_members_team_situation |
| 4563 | half - head - biggest - write - excited | 22 | 4563_half_head_biggest_write |
| 4564 | special - reply - example - luck - utc | 22 | 4564_special_reply_example_luck |
| 4565 | smart - services - blockchain - transaction - web | 22 | 4565_smart_services_blockchain_transaction |
| 4566 | links - free - group - link - | 22 | 4566_links_free_group_link |
| 4567 | active - - - - | 22 | 4567_active___ |
| 4568 | words - brother - little - problem - guy | 22 | 4568_words_brother_little_problem |
| 4569 | better - person - bad - moment - trust | 22 | 4569_better_person_bad_moment |
| 4570 | - - - - | 22 | 4570____ |
| 4571 | rest - happy - good - week - home | 22 | 4571_rest_happy_good_week |
| 4572 | message - fun - chat - real - | 22 | 4572_message_fun_chat_real |
| 4573 | difference - face - correct - nice - question | 22 | 4573_difference_face_correct_nice |
| 4574 | fake - attention - links - private - accounts | 22 | 4574_fake_attention_links_private |
| 4575 | marketing - - - - | 22 | 4575_marketing___ |
| 4576 | reply - upcoming - stuff - chain - old | 22 | 4576_reply_upcoming_stuff_chain |
| 4577 | bot - try - best - - | 22 | 4577_bot_try_best_ |
| 4578 | chain - step - article - interested - rest | 22 | 4578_chain_step_article_interested |
| 4579 | situation - hour - friend - nice - right | 22 | 4579_situation_hour_friend_nice |
| 4580 | financial - hand - data - control - major | 22 | 4580_financial_hand_data_control |
| 4581 | number - daily - - - | 22 | 4581_number_daily__ |
| 4582 | financial - head - trade - type - account | 22 | 4582_financial_head_trade_type |
| 4583 | account - pm - official - link - | 22 | 4583_account_pm_official_link |
| 4584 | security - ethereum - specific - current - news | 22 | 4584_security_ethereum_specific_current |
| 4585 | bit - course - strong - person - problem | 22 | 4585_bit_course_strong_person |
| 4586 | official - problem - kind - bit - step | 22 | 4586_official_problem_kind_bit |
| 4587 | email - phone - contact - accounts - services | 22 | 4587_email_phone_contact_accounts |
| 4588 | channel - account - ask - dm - | 22 | 4588_channel_account_ask_dm |
| 4589 | huge - report - country - years - holders | 22 | 4589_huge_report_country_years |
| 4590 | country - digital - assets - transactions - payment | 22 | 4590_country_digital_assets_transactions |
| 4591 | latest - financial - access - support - similar | 22 | 4591_latest_financial_access_support |
| 4592 | market - today - good - morning - price | 22 | 4592_market_today_good_morning |
| 4593 | crypto - option - country - term - huge | 22 | 4593_crypto_option_country_term |
| 4594 | biggest - attention - bro - job - problem | 22 | 4594_biggest_attention_bro_job |
| 4595 | current - tomorrow - wait - low - | 22 | 4595_current_tomorrow_wait_low |
| 4596 | little - worth - fee - marketing - lol | 22 | 4596_little_worth_fee_marketing |
| 4597 | normal - problems - problem - special - situation | 22 | 4597_normal_problems_problem_special |
| 4598 | major - features - exciting - article - future | 22 | 4598_major_features_exciting_article |
| 4599 | step - assets - ethereum - power - website | 22 | 4599_step_assets_ethereum_power |
| 4600 | sense - correct - words - hour - point | 22 | 4600_sense_correct_words_hour |
| 4601 | special - past - legit - experience - platform | 22 | 4601_special_past_legit_experience |
| 4602 | new - - - - | 22 | 4602_new___ |
| 4603 | gone - opinion - normal - correct - members | 22 | 4603_gone_opinion_normal_correct |
| 4604 | half - better - free - related - | 22 | 4604_half_better_free_related |
| 4605 | users - - - - | 22 | 4605_users___ |
| 4606 | sorry - close - yesterday - shit - alot | 22 | 4606_sorry_close_yesterday_shit |
| 4607 | exciting - head - release - option - home | 22 | 4607_exciting_head_release_option |
| 4608 | little - dm - - - | 22 | 4608_little_dm__ |
| 4609 | hand - bitcoin - current - idea - country | 22 | 4609_hand_bitcoin_current_idea |
| 4610 | power - minutes - hours - - | 22 | 4610_power_minutes_hours_ |
| 4611 | crypto - country - rules - article - read | 22 | 4611_crypto_country_rules_article |
| 4612 | hours - new - - - | 22 | 4612_hours_new__ |
| 4613 | interesting - period - short - - | 22 | 4613_interesting_period_short_ |
| 4614 | token - bad - news - transaction - pepe | 22 | 4614_token_bad_news_transaction |
| 4615 | interested - price - game - man - end | 22 | 4615_interested_price_game_man |
| 4616 | major - features - ai - latest - new | 22 | 4616_major_features_ai_latest |
| 4617 | potential - way - great - positive - public | 22 | 4617_potential_way_great_positive |
| 4618 | short - small - little - spot - order | 22 | 4618_short_small_little_spot |
| 4619 | ai - - - - | 22 | 4619_ai___ |
| 4620 | biggest - glad - alot - personal - users | 22 | 4620_biggest_glad_alot_personal |
| 4621 | defi - bot - welcome - opportunity - world | 22 | 4621_defi_bot_welcome_opportunity |
| 4622 | opportunity - try - life - future - happy | 22 | 4622_opportunity_try_life_future |
| 4623 | position - chart - holders - funds - buy | 22 | 4623_position_chart_holders_funds |
| 4624 | single - different - thing - rest - mind | 22 | 4624_single_different_thing_rest |
| 4625 | face - guy - type - person - open | 22 | 4625_face_guy_type_person |
| 4626 | multiple - ethereum - chain - network - support | 22 | 4626_multiple_ethereum_chain_network |
| 4627 | true - likely - man - nice - hours | 22 | 4627_true_likely_man_nice |
| 4628 | users - - - - | 22 | 4628_users___ |
| 4629 | admin - specific - action - close - guy | 22 | 4629_admin_specific_action_close |
| 4630 | mind - people - short - time - | 22 | 4630_mind_people_short_time |
| 4631 | tho - question - ones - wish - simple | 22 | 4631_tho_question_ones_wish |
| 4632 | fees - online - happy - interested - investing | 22 | 4632_fees_online_happy_interested |
| 4633 | unique - different - check - bitcoin - rate | 22 | 4633_unique_different_check_bitcoin |
| 4634 | projects - clear - option - certain - brother | 22 | 4634_projects_clear_option_certain |
| 4635 | job - drop - course - pm - time | 22 | 4635_job_drop_course_pm |
| 4636 | interested - crazy - bad - ones - possible | 22 | 4636_interested_crazy_bad_ones |
| 4637 | accounts - note - twitter - announcement - period | 22 | 4637_accounts_note_twitter_announcement |
| 4638 | guy - shit - fine - person - means | 22 | 4638_guy_shit_fine_person |
| 4639 | ai - code - likely - important - public | 22 | 4639_ai_code_likely_important |
| 4640 | trading - sense - announcement - day - | 22 | 4640_trading_sense_announcement_day |
| 4641 | report - weeks - recent - months - plans | 22 | 4641_report_weeks_recent_months |
| 4642 | bad - news - certain - market - click | 22 | 4642_bad_news_certain_market |
| 4643 | situation - doubt - country - current - sorry | 22 | 4643_situation_doubt_country_current |
| 4644 | help - good - - - | 22 | 4644_help_good__ |
| 4645 | site - small - legit - video - online | 22 | 4645_site_small_legit_video |
| 4646 | legit - post - real - thanks - people | 22 | 4646_legit_post_real_thanks |
| 4647 | couple - bad - normal - glad - low | 22 | 4647_couple_bad_normal_glad |
| 4648 | cause - happy - trade - lot - profit | 22 | 4648_cause_happy_trade_lot |
| 4649 | global - play - family - country - games | 22 | 4649_global_play_family_country |
| 4650 | active - - - - | 22 | 4650_active___ |
| 4651 | single - fine - power - course - years | 22 | 4651_single_fine_power_course |
| 4652 | late - luck - idea - possible - opportunity | 22 | 4652_late_luck_idea_possible |
| 4653 | hand - dude - interesting - space - important | 22 | 4653_hand_dude_interesting_space |
| 4654 | action - link - - - | 22 | 4654_action_link__ |
| 4655 | latest - user - users - complete - cryptocurrency | 22 | 4655_latest_user_users_complete |
| 4656 | nice - kind - mate - little - today | 22 | 4656_nice_kind_mate_little |
| 4657 | - - - - | 22 | 4657____ |
| 4658 | main - protocol - feature - ai - word | 22 | 4658_main_protocol_feature_ai |
| 4659 | data - action - - - | 22 | 4659_data_action__ |
| 4660 | security - ai - need - minutes - watch | 22 | 4660_security_ai_need_minutes |
| 4661 | network - course - transaction - transactions - bitcoin | 22 | 4661_network_course_transaction_transactions |
| 4662 | nft - project - idea - strong - token | 22 | 4662_nft_project_idea_strong |
| 4663 | control - protocol - web - ecosystem - app | 22 | 4663_control_protocol_web_ecosystem |
| 4664 | price - rules - luck - period - multiple | 22 | 4664_price_rules_luck_period |
| 4665 | reward - transaction - days - - | 22 | 4665_reward_transaction_days_ |
| 4666 | rules - number - - - | 22 | 4666_rules_number__ |
| 4667 | dev - services - contact - post - need | 22 | 4667_dev_services_contact_post |
| 4668 | tokens - period - site - token - old | 22 | 4668_tokens_period_site_token |
| 4669 | value - marketing - support - features - exciting | 22 | 4669_value_marketing_support_features |
| 4670 | today - good - team - fine - job | 22 | 4670_today_good_team_fine |
| 4671 | tokens - couple - rules - situation - token | 22 | 4671_tokens_couple_rules_situation |
| 4672 | coins - fine - couple - coin - holders | 22 | 4672_coins_fine_couple_coin |
| 4673 | spot - situation - order - security - code | 22 | 4673_spot_situation_order_security |
| 4674 | pls - fees - fee - hour - chart | 22 | 4674_pls_fees_fee_hour |
| 4675 | visit - games - unique - nft - note | 22 | 4675_visit_games_unique_nft |
| 4676 | guys - bit - minutes - point - questions | 22 | 4676_guys_bit_minutes_point |
| 4677 | tho - friend - guy - main - bit | 22 | 4677_tho_friend_guy_main |
| 4678 | global - members - community - meet - love | 22 | 4678_global_members_community_meet |
| 4679 | update - community - rest - sense - difference | 22 | 4679_update_community_rest_sense |
| 4680 | follow - updates - day - friends - set | 22 | 4680_follow_updates_day_friends |
| 4681 | potential - price - months - recent - market | 22 | 4681_potential_price_months_recent |
| 4682 | token - term - tokens - interesting - holders | 21 | 4682_token_term_tokens_interesting |
| 4683 | rules - doubt - research - report - binance | 21 | 4683_rules_doubt_research_report |
| 4684 | digital - price - - - | 21 | 4684_digital_price__ |
| 4685 | smart - platform - game - - | 21 | 4685_smart_platform_game_ |
| 4686 | deal - higher - selling - news - price | 21 | 4686_deal_higher_selling_news |
| 4687 | ai - space - twitter - live - data | 21 | 4687_ai_space_twitter_live |
| 4688 | site - form - members - games - research | 21 | 4688_site_form_members_games |
| 4689 | binance - network - btc - bitcoin - similar | 21 | 4689_binance_network_btc_bitcoin |
| 4690 | control - buying - financial - option - money | 21 | 4690_control_buying_financial_option |
| 4691 | account - control - previous - weeks - problem | 21 | 4691_account_control_previous_weeks |
| 4692 | key - fix - close - issue - problem | 21 | 4692_key_fix_close_issue |
| 4693 | dev - shit - lot - strong - person | 21 | 4693_dev_shit_lot_strong |
| 4694 | main - lots - private - telegram - members | 21 | 4694_main_lots_private_telegram |
| 4695 | product - early - couple - public - release | 21 | 4695_product_early_couple_public |
| 4696 | cool - right - free - - | 21 | 4696_cool_right_free_ |
| 4697 | search - feature - option - opinion - users | 21 | 4697_search_feature_option_opinion |
| 4698 | site - drop - transactions - claim - information | 21 | 4698_site_drop_transactions_claim |
| 4699 | post - group - free - new - | 21 | 4699_post_group_free_new |
| 4700 | services - country - main - course - selling | 21 | 4700_services_country_main_course |
| 4701 | latest - updates - important - news - information | 21 | 4701_latest_updates_important_news |
| 4702 | price - moment - big - - | 21 | 4702_price_moment_big_ |
| 4703 | defi - funds - option - multiple - point | 21 | 4703_defi_funds_option_multiple |
| 4704 | months - crazy - btc - small - coin | 21 | 4704_months_crazy_btc_small |
| 4705 | guy - face - interesting - word - deal | 21 | 4705_guy_face_interesting_word |
| 4706 | problem - process - rewards - - | 21 | 4706_problem_process_rewards_ |
| 4707 | - - - - | 21 | 4707____ |
| 4708 | word - issues - fun - reason - nice | 21 | 4708_word_issues_fun_reason |
| 4709 | report - projects - information - exchange - fake | 21 | 4709_report_projects_information_exchange |
| 4710 | fun - day - hour - times - hours | 21 | 4710_fun_day_hour_times |
| 4711 | price - bitcoin - - - | 21 | 4711_price_bitcoin__ |
| 4712 | past - pay - app - technology - recent | 21 | 4712_past_pay_app_technology |
| 4713 | market - best - volume - assets - risk | 21 | 4713_market_best_volume_assets |
| 4714 | users - private - - - | 21 | 4714_users_private__ |
| 4715 | ecosystem - attention - damn - near - dude | 21 | 4715_ecosystem_attention_damn_near |
| 4716 | previous - chart - close - correct - second | 21 | 4716_previous_chart_close_correct |
| 4717 | users - payment - - - | 21 | 4717_users_payment__ |
| 4718 | business - profit - week - investors - days | 21 | 4718_business_profit_week_investors |
| 4719 | massive - updates - day - price - | 21 | 4719_massive_updates_day_price |
| 4720 | sir - active - mate - channel - friend | 21 | 4720_sir_active_mate_channel |
| 4721 | second - digital - public - successful - plans | 21 | 4721_second_digital_public_successful |
| 4722 | link - - - - | 21 | 4722_link___ |
| 4723 | article - nft - higher - certain - second | 21 | 4723_article_nft_higher_certain |
| 4724 | interesting - important - things - protocol - different | 21 | 4724_interesting_important_things_protocol |
| 4725 | hard - things - today - time - fact | 21 | 4725_hard_things_today_time |
| 4726 | dude - sense - special - showing - past | 21 | 4726_dude_sense_special_showing |
| 4727 | dear - game - play - - | 21 | 4727_dear_game_play_ |
| 4728 | fees - withdrawal - special - online - real | 21 | 4728_fees_withdrawal_special_online |
| 4729 | chat - dm - main - lol - place | 21 | 4729_chat_dm_main_lol |
| 4730 | earlier - couple - fact - correct - morning | 21 | 4730_earlier_couple_fact_correct |
| 4731 | trust - ethereum - - - | 21 | 4731_trust_ethereum__ |
| 4732 | bitcoin - btc - market - price - limited | 21 | 4732_bitcoin_btc_market_price |
| 4733 | matter - opinion - difference - things - point | 21 | 4733_matter_opinion_difference_things |
| 4734 | article - recent - fun - business - course | 21 | 4734_article_recent_fun_business |
| 4735 | possible - market - global - today - second | 21 | 4735_possible_market_global_today |
| 4736 | past - development - blockchain - future - network | 21 | 4736_past_development_blockchain_future |
| 4737 | man - possible - place - - | 21 | 4737_man_possible_place_ |
| 4738 | interested - message - lots - dm - things | 21 | 4738_interested_message_lots_dm |
| 4739 | global - rules - digital - control - welcome | 21 | 4739_global_rules_digital_control |
| 4740 | situation - correct - problem - personal - search | 21 | 4740_situation_correct_problem_personal |
| 4741 | step - opinion - man - short - time | 21 | 4741_step_opinion_man_short |
| 4742 | hand - minutes - tomorrow - dear - week | 21 | 4742_hand_minutes_tomorrow_dear |
| 4743 | friends - updates - telegram - space - update | 21 | 4743_friends_updates_telegram_space |
| 4744 | thinking - weeks - bad - period - minutes | 21 | 4744_thinking_weeks_bad_period |
| 4745 | business - care - damn - country - hour | 21 | 4745_business_care_damn_country |
| 4746 | fact - important - talk - worth - place | 21 | 4746_fact_important_talk_worth |
| 4747 | order - look - difference - way - support | 21 | 4747_order_look_difference_way |
| 4748 | pepe - exchange - major - news - email | 21 | 4748_pepe_exchange_major_news |
| 4749 | issue - minutes - data - utc - service | 21 | 4749_issue_minutes_data_utc |
| 4750 | near - likely - focus - term - action | 21 | 4750_near_likely_focus_term |
| 4751 | listing - exchange - contact - half - admin | 21 | 4751_listing_exchange_contact_half |
| 4752 | digital - blockchain - technology - moment - multiple | 21 | 4752_digital_blockchain_technology_moment |
| 4753 | situation - chain - likely - specific - assets | 21 | 4753_situation_chain_likely_specific |
| 4754 | site - chain - details - active - tokens | 21 | 4754_site_chain_details_active |
| 4755 | month - trade - free - money - | 21 | 4755_month_trade_free_money |
| 4756 | plans - fee - withdrawal - investment - best | 21 | 4756_plans_fee_withdrawal_investment |
| 4757 | couple - brother - sorry - reason - important | 21 | 4757_couple_brother_sorry_reason |
| 4758 | late - spot - post - available - total | 21 | 4758_late_spot_post_available |
| 4759 | different - face - interesting - kind - thing | 21 | 4759_different_face_interesting_kind |
| 4760 | minimum - deposit - hours - withdrawal - new | 21 | 4760_minimum_deposit_hours_withdrawal |
| 4761 | home - half - opinion - awesome - buddy | 21 | 4761_home_half_opinion_awesome |
| 4762 | times - current - digital - years - lot | 21 | 4762_times_current_digital_years |
| 4763 | personal - address - moment - deposit - network | 21 | 4763_personal_address_moment_deposit |
| 4764 | reason - user - message - word - new | 21 | 4764_reason_user_message_word |
| 4765 | perfect - chance - person - special - fact | 21 | 4765_perfect_chance_person_special |
| 4766 | plan - level - daily - invest - trust | 21 | 4766_plan_level_daily_invest |
| 4767 | feature - matter - key - point - update | 21 | 4767_feature_matter_key_point |
| 4768 | telegram - - - - | 21 | 4768_telegram___ |
| 4769 | research - plan - access - twitter - smart | 21 | 4769_research_plan_access_twitter |
| 4770 | previous - launch - huge - contract - rewards | 21 | 4770_previous_launch_huge_contract |
| 4771 | pool - community - - - | 21 | 4771_pool_community__ |
| 4772 | minimum - term - fee - account - latest | 21 | 4772_minimum_term_fee_account |
| 4773 | tomorrow - life - correct - course - morning | 21 | 4773_tomorrow_life_correct_course |
| 4774 | positive - strong - hold - power - lots | 21 | 4774_positive_strong_hold_power |
| 4775 | issues - transactions - blockchain - bitcoin - | 21 | 4775_issues_transactions_blockchain_bitcoin |
| 4776 | idea - interesting - interested - worth - lol | 21 | 4776_idea_interesting_interested_worth |
| 4777 | report - block - clear - ways - option | 21 | 4777_report_block_clear_ways |
| 4778 | night - plan - friend - family - man | 21 | 4778_night_plan_friend_family |
| 4779 | link - fun - today - - | 21 | 4779_link_fun_today_ |
| 4780 | active - support - - - | 21 | 4780_active_support__ |
| 4781 | price - rate - hard - listing - market | 21 | 4781_price_rate_hard_listing |
| 4782 | version - ethereum - fact - bitcoin - long | 21 | 4782_version_ethereum_fact_bitcoin |
| 4783 | ecosystem - drop - announcement - upcoming - normal | 21 | 4783_ecosystem_drop_announcement_upcoming |
| 4784 | nice - buddy - close - word - kind | 21 | 4784_nice_buddy_close_word |
| 4785 | certain - brother - point - chart - tho | 21 | 4785_certain_brother_point_chart |
| 4786 | crazy - person - swap - rewards - guys | 21 | 4786_crazy_person_swap_rewards |
| 4787 | simple - home - problems - life - happy | 21 | 4787_simple_home_problems_life |
| 4788 | dev - couple - word - person - true | 21 | 4788_dev_couple_word_person |
| 4789 | official - wallet - channel - token - team | 21 | 4789_official_wallet_channel_token |
| 4790 | risk - profit - trading - luck - huge | 21 | 4790_risk_profit_trading_luck |
| 4791 | issue - sense - matter - step - latest | 21 | 4791_issue_sense_matter_step |
| 4792 | daily - deposit - fee - profit - day | 21 | 4792_daily_deposit_fee_profit |
| 4793 | investment - - - - | 21 | 4793_investment___ |
| 4794 | site - website - web - multiple - official | 21 | 4794_site_website_web_multiple |
| 4795 | deal - sure - weeks - phone - email | 21 | 4795_deal_sure_weeks_phone |
| 4796 | major - cryptocurrency - public - business - members | 21 | 4796_major_cryptocurrency_public_business |
| 4797 | country - accounts - list - ready - business | 21 | 4797_country_accounts_list_ready |
| 4798 | event - bitcoin - safe - read - | 21 | 4798_event_bitcoin_safe_read |
| 4799 | data - fine - type - user - things | 21 | 4799_data_fine_type_user |
| 4800 | strong - project - ready - successful - fun | 21 | 4800_strong_project_ready_successful |
| 4801 | protocol - swap - site - news - pay | 21 | 4801_protocol_swap_site_news |
| 4802 | fee - fees - token - tokens - claim | 21 | 4802_fee_fees_token_tokens |
| 4803 | nft - lots - different - chain - value | 21 | 4803_nft_lots_different_chain |
| 4804 | global - assets - - - | 21 | 4804_global_assets__ |
| 4805 | shit - alot - lol - amazing - online | 21 | 4805_shit_alot_lol_amazing |
| 4806 | protocol - great - new - - | 21 | 4806_protocol_great_new_ |
| 4807 | normal - clear - ta - example - tho | 21 | 4807_normal_clear_ta_example |
| 4808 | brother - cool - personal - step - little | 21 | 4808_brother_cool_personal_step |
| 4809 | years - close - crazy - interesting - period | 21 | 4809_years_close_crazy_interesting |
| 4810 | issue - old - issues - fix - solution | 21 | 4810_issue_old_issues_fix |
| 4811 | little - life - excited - exciting - lots | 21 | 4811_little_life_excited_exciting |
| 4812 | weeks - problems - bit - coin - lots | 21 | 4812_weeks_problems_bit_coin |
| 4813 | members - message - guys - latest - ama | 21 | 4813_members_message_guys_latest |
| 4814 | guide - web - app - important - trading | 21 | 4814_guide_web_app_important |
| 4815 | exchange - - - - | 21 | 4815_exchange___ |
| 4816 | today - questions - question - works - | 21 | 4816_today_questions_question_works |
| 4817 | interested - late - pm - guys - day | 21 | 4817_interested_late_pm_guys |
| 4818 | quick - welcome - day - message - dm | 21 | 4818_quick_welcome_day_message |
| 4819 | true - active - - - | 21 | 4819_true_active__ |
| 4820 | projects - past - experience - new - | 21 | 4820_projects_past_experience_new |
| 4821 | comment - face - crazy - guy - lots | 21 | 4821_comment_face_crazy_guy |
| 4822 | position - exchange - control - projects - similar | 21 | 4822_position_exchange_control_projects |
| 4823 | ai - future - higher - ways - thinking | 21 | 4823_ai_future_higher_ways |
| 4824 | - - - - | 21 | 4824____ |
| 4825 | - - - - | 21 | 4825____ |
| 4826 | risk - way - financial - services - market | 21 | 4826_risk_way_financial_services |
| 4827 | main - - - - | 21 | 4827_main___ |
| 4828 | second - end - year - place - lot | 21 | 4828_second_end_year_place |
| 4829 | funds - exchange - cryptocurrency - - | 21 | 4829_funds_exchange_cryptocurrency_ |
| 4830 | release - higher - data - bitcoin - near | 21 | 4830_release_higher_data_bitcoin |
| 4831 | liquidity - things - week - bitcoin - fact | 21 | 4831_liquidity_things_week_bitcoin |
| 4832 | tg - bot - use - rules - chat | 21 | 4832_tg_bot_use_rules |
| 4833 | fake - account - crypto - words - coin | 21 | 4833_fake_account_crypto_words |
| 4834 | difference - win - small - wrong - order | 21 | 4834_difference_win_small_wrong |
| 4835 | public - volume - number - true - rewards | 21 | 4835_public_volume_number_true |
| 4836 | good - alot - mate - active - work | 21 | 4836_good_alot_mate_active |
| 4837 | little - amazing - try - start - help | 21 | 4837_little_amazing_try_start |
| 4838 | news - social - bit - second - little | 21 | 4838_news_social_bit_second |
| 4839 | updates - development - ama - official - latest | 21 | 4839_updates_development_ama_official |
| 4840 | hard - friend - guy - friends - job | 21 | 4840_hard_friend_guy_friends |
| 4841 | face - mind - thing - game - time | 21 | 4841_face_mind_thing_game |
| 4842 | - - - - | 21 | 4842____ |
| 4843 | special - fact - past - interesting - shows | 21 | 4843_special_fact_past_interesting |
| 4844 | bot - service - moment - thing - | 21 | 4844_bot_service_moment_thing |
| 4845 | reply - guy - months - true - man | 21 | 4845_reply_guy_months_true |
| 4846 | work - guys - head - good - rewards | 21 | 4846_work_guys_head_good |
| 4847 | issues - certain - network - shit - night | 21 | 4847_issues_certain_network_shit |
| 4848 | market - liquidity - value - big - start | 21 | 4848_market_liquidity_value_big |
| 4849 | deal - massive - trust - wallet - btc | 21 | 4849_deal_massive_trust_wallet |
| 4850 | dude - couple - deal - man - live | 21 | 4850_dude_couple_deal_man |
| 4851 | public - information - official - launch - special | 21 | 4851_public_information_official_launch |
| 4852 | links - message - potential - funds - wallet | 21 | 4852_links_message_potential_funds |
| 4853 | - - - - | 21 | 4853____ |
| 4854 | pepe - projects - project - solution - upcoming | 21 | 4854_pepe_projects_project_solution |
| 4855 | takes - upcoming - minutes - version - tokens | 21 | 4855_takes_upcoming_minutes_version |
| 4856 | quick - higher - rate - minutes - service | 21 | 4856_quick_higher_rate_minutes |
| 4857 | multiple - low - ama - high - launch | 21 | 4857_multiple_low_ama_high |
| 4858 | dude - certain - biggest - investors - crazy | 21 | 4858_dude_certain_biggest_investors |
| 4859 | pepe - date - note - coin - reward | 21 | 4859_pepe_date_note_coin |
| 4860 | product - new - upcoming - release - alot | 21 | 4860_product_new_upcoming_release |
| 4861 | new - time - good - - | 21 | 4861_new_time_good_ |
| 4862 | answer - buddy - deal - wrong - sorry | 21 | 4862_answer_buddy_deal_wrong |
| 4863 | strong - perfect - guys - - | 21 | 4863_strong_perfect_guys_ |
| 4864 | rules - positive - members - dear - questions | 21 | 4864_rules_positive_members_dear |
| 4865 | sense - brother - example - cool - place | 21 | 4865_sense_brother_example_cool |
| 4866 | online - reason - person - business - safe | 21 | 4866_online_reason_person_business |
| 4867 | situation - assets - rest - digital - certain | 21 | 4867_situation_assets_rest_digital |
| 4868 | deal - crypto - likely - assets - risk | 21 | 4868_deal_crypto_likely_assets |
| 4869 | special - search - smart - country - months | 21 | 4869_special_search_smart_country |
| 4870 | hand - solution - head - country - visit | 21 | 4870_hand_solution_head_country |
| 4871 | welcome - fun - group - chat - guys | 21 | 4871_welcome_fun_group_chat |
| 4872 | reward - event - sir - trading - days | 21 | 4872_reward_event_sir_trading |
| 4873 | problems - payment - shit - digital - transaction | 21 | 4873_problems_payment_shit_digital |
| 4874 | product - fact - long - works - years | 21 | 4874_product_fact_long_works |
| 4875 | position - chart - holders - type - funds | 21 | 4875_position_chart_holders_type |
| 4876 | massive - marketing - low - contract - experience | 21 | 4876_massive_marketing_low_contract |
| 4877 | app - time - sir - process - experience | 21 | 4877_app_time_sir_process |
| 4878 | glad - welcome - mate - awesome - happy | 21 | 4878_glad_welcome_mate_awesome |
| 4879 | key - claim - btc - price - bitcoin | 21 | 4879_key_claim_btc_price |
| 4880 | contact - days - time - available - accounts | 21 | 4880_contact_days_time_available |
| 4881 | best - - - - | 20 | 4881_best___ |
| 4882 | total - answer - difference - correct - hard | 20 | 4882_total_answer_difference_correct |
| 4883 | idea - old - alot - night - type | 20 | 4883_idea_old_alot_night |
| 4884 | bitcoin - price - couple - late - recent | 20 | 4884_bitcoin_price_couple_late |
| 4885 | sir - funds - glad - yesterday - past | 20 | 4885_sir_funds_glad_yesterday |
| 4886 | increase - little - bit - market - today | 20 | 4886_increase_little_bit_market |
| 4887 | ready - extra - level - try - investors | 20 | 4887_ready_extra_level_try |
| 4888 | luck - good - - - | 20 | 4888_luck_good__ |
| 4889 | ai - trading - bot - technology - experience | 20 | 4889_ai_trading_bot_technology |
| 4890 | marketing - technology - user - features - data | 20 | 4890_marketing_technology_user_features |
| 4891 | users - rules - order - user - projects | 20 | 4891_users_rules_order_user |
| 4892 | normal - lol - shit - alot - guy | 20 | 4892_normal_lol_shit_alot |
| 4893 | country - fair - complete - private - world | 20 | 4893_country_fair_complete_private |
| 4894 | course - - - - | 20 | 4894_course___ |
| 4895 | pepe - launch - times - liquidity - contract | 20 | 4895_pepe_launch_times_liquidity |
| 4896 | details - - - - | 20 | 4896_details___ |
| 4897 | fun - yesterday - guy - ones - lol | 20 | 4897_fun_yesterday_guy_ones |
| 4898 | details - thanks - message - help - community | 20 | 4898_details_thanks_message_help |
| 4899 | cool - data - fact - smart - projects | 20 | 4899_cool_data_fact_smart |
| 4900 | strong - chance - end - - | 20 | 4900_strong_chance_end_ |
| 4901 | personal - difference - ones - future - attention | 20 | 4901_personal_difference_ones_future |
| 4902 | ama - project - projects - brother - opinion | 20 | 4902_ama_project_projects_brother |
| 4903 | real - technology - web - guide - links | 20 | 4903_real_technology_web_guide |
| 4904 | marketing - ready - announcement - latest - article | 20 | 4904_marketing_ready_announcement_latest |
| 4905 | phone - following - number - email - transaction | 20 | 4905_phone_following_number_email |
| 4906 | single - chat - near - secure - pretty | 20 | 4906_single_chat_near_secure |
| 4907 | control - situation - words - problems - sense | 20 | 4907_control_situation_words_problems |
| 4908 | friends - real - network - accounts - hard | 20 | 4908_friends_real_network_accounts |
| 4909 | user - liquidity - business - ways - fair | 20 | 4909_user_liquidity_business_ways |
| 4910 | term - minimum - rate - short - legit | 20 | 4910_term_minimum_rate_short |
| 4911 | - - - - | 20 | 4911____ |
| 4912 | tho - mean - friend - video - year | 20 | 4912_tho_mean_friend_video |
| 4913 | buying - - - - | 20 | 4913_buying___ |
| 4914 | single - chance - - - | 20 | 4914_single_chance__ |
| 4915 | massive - marketing - fair - strong - fee | 20 | 4915_massive_marketing_fair_strong |
| 4916 | word - night - better - read - position | 20 | 4916_word_night_better_read |
| 4917 | post - number - details - available - issues | 20 | 4917_post_number_details_available |
| 4918 | buddy - mate - dude - sure - answer | 20 | 4918_buddy_mate_dude_sure |
| 4919 | digital - year - days - group - investment | 20 | 4919_digital_year_days_group |
| 4920 | - - - - | 20 | 4920____ |
| 4921 | short - break - weeks - idea - months | 20 | 4921_short_break_weeks_idea |
| 4922 | link - wallet - - - | 20 | 4922_link_wallet__ |
| 4923 | ethereum - question - place - fees - better | 20 | 4923_ethereum_question_place_fees |
| 4924 | research - case - - - | 20 | 4924_research_case__ |
| 4925 | fix - quick - deal - page - available | 20 | 4925_fix_quick_deal_page |
| 4926 | list - - - - | 20 | 4926_list___ |
| 4927 | likely - latest - opinion - idea - case | 20 | 4927_likely_latest_opinion_idea |
| 4928 | tokens - network - address - volume - clear | 20 | 4928_tokens_network_address_volume |
| 4929 | coin - days - new - hours - | 20 | 4929_coin_days_new_hours |
| 4930 | assets - members - group - global - solution | 20 | 4930_assets_members_group_global |
| 4931 | issues - chain - block - special - minutes | 20 | 4931_issues_chain_block_special |
| 4932 | btc - months - private - information - month | 20 | 4932_btc_months_private_information |
| 4933 | smart - news - - - | 20 | 4933_smart_news__ |
| 4934 | deal - cryptocurrency - bitcoin - wallet - | 20 | 4934_deal_cryptocurrency_bitcoin_wallet |
| 4935 | cryptocurrency - interested - group - trading - link | 20 | 4935_cryptocurrency_interested_group_trading |
| 4936 | total - members - community - - | 20 | 4936_total_members_community_ |
| 4937 | special - normal - reply - phone - man | 20 | 4937_special_normal_reply_phone |
| 4938 | - - - - | 20 | 4938____ |
| 4939 | community - - - - | 20 | 4939_community___ |
| 4940 | friends - rewards - value - - | 20 | 4940_friends_rewards_value_ |
| 4941 | morning - easy - friends - life - good | 20 | 4941_morning_easy_friends_life |
| 4942 | video - cryptocurrency - new - - | 20 | 4942_video_cryptocurrency_new_ |
| 4943 | free - luck - minimum - withdraw - announcement | 20 | 4943_free_luck_minimum_withdraw |
| 4944 | recent - transactions - fees - network - community | 20 | 4944_recent_transactions_fees_network |
| 4945 | - - - - | 20 | 4945____ |
| 4946 | contract - old - - - | 20 | 4946_contract_old__ |
| 4947 | participate - details - sign - win - utc | 20 | 4947_participate_details_sign_win |
| 4948 | coins - minimum - coin - network - tokens | 20 | 4948_coins_minimum_coin_network |
| 4949 | fair - dev - launch - personal - liquidity | 20 | 4949_fair_dev_launch_personal |
| 4950 | home - phone - simple - ones - smart | 20 | 4950_home_phone_simple_ones |
| 4951 | data - important - specific - main - sure | 20 | 4951_data_important_specific_main |
| 4952 | crazy - opinion - shit - fine - spot | 20 | 4952_crazy_opinion_shit_fine |
| 4953 | normal - reply - game - bro - lol | 20 | 4953_normal_reply_game_bro |
| 4954 | current - - - - | 20 | 4954_current___ |
| 4955 | supply - total - showing - correct - different | 20 | 4955_supply_total_showing_correct |
| 4956 | moment - week - year - okay - tomorrow | 20 | 4956_moment_week_year_okay |
| 4957 | fun - single - man - real - chat | 20 | 4957_fun_single_man_real |
| 4958 | questions - ai - wrong - power - day | 20 | 4958_questions_ai_wrong_power |
| 4959 | app - withdraw - global - account - code | 20 | 4959_app_withdraw_global_account |
| 4960 | discord - protocol - digital - media - upcoming | 20 | 4960_discord_protocol_digital_media |
| 4961 | weeks - end - btc - price - bitcoin | 20 | 4961_weeks_end_btc_price |
| 4962 | sense - recent - technology - form - multiple | 20 | 4962_sense_recent_technology_form |
| 4963 | legit - profit - link - fee - pay | 20 | 4963_legit_profit_link_fee |
| 4964 | service - - - - | 20 | 4964_service___ |
| 4965 | recent - - - - | 20 | 4965_recent___ |
| 4966 | services - telegram - service - new - web | 20 | 4966_services_telegram_service_new |
| 4967 | term - running - deposit - date - short | 20 | 4967_term_running_deposit_date |
| 4968 | fees - late - little - fee - earn | 20 | 4968_fees_late_little_fee |
| 4969 | spot - single - yesterday - weeks - upcoming | 20 | 4969_spot_single_yesterday_weeks |
| 4970 | clear - simple - note - business - rate | 20 | 4970_clear_simple_note_business |
| 4971 | users - attention - matter - public - links | 20 | 4971_users_attention_matter_public |
| 4972 | nice - day - community - today - | 20 | 4972_nice_day_community_today |
| 4973 | global - exchange - binance - - | 20 | 4973_global_exchange_binance_ |
| 4974 | page - active - fast - big - | 20 | 4974_page_active_fast_big |
| 4975 | bitcoin - chance - account - click - | 20 | 4975_bitcoin_chance_account_click |
| 4976 | ai - data - high - talk - different | 20 | 4976_ai_data_high_talk |
| 4977 | tho - fair - great - pay - thanks | 20 | 4977_tho_fair_great_pay |
| 4978 | project - projects - single - research - early | 20 | 4978_project_projects_single_research |
| 4979 | reason - dude - head - ones - position | 20 | 4979_reason_dude_head_ones |
| 4980 | assets - value - crypto - claim - coins | 20 | 4980_assets_value_crypto_claim |
| 4981 | business - - - - | 20 | 4981_business___ |
| 4982 | trust - users - fact - possible - idea | 20 | 4982_trust_users_fact_possible |
| 4983 | opinion - guys - nice - game - sure | 20 | 4983_opinion_guys_nice_game |
| 4984 | action - link - games - links - admin | 20 | 4984_action_link_games_links |
| 4985 | tomorrow - days - today - day - | 20 | 4985_tomorrow_days_today_day |
| 4986 | recent - digital - - - | 20 | 4986_recent_digital__ |
| 4987 | pm - message - single - fun - dm | 20 | 4987_pm_message_single_fun |
| 4988 | coins - key - control - example - coin | 20 | 4988_coins_key_control_example |
| 4989 | - - - - | 20 | 4989____ |
| 4990 | stuff - problem - update - issue - simple | 20 | 4990_stuff_problem_update_issue |
| 4991 | - - - - | 20 | 4991____ |
| 4992 | huge - wait - digital - order - article | 20 | 4992_huge_wait_digital_order |
| 4993 | latest - capital - bitcoin - product - ways | 20 | 4993_latest_capital_bitcoin_product |
| 4994 | payment - clear - stuff - option - digital | 20 | 4994_payment_clear_stuff_option |
| 4995 | focus - search - past - short - success | 20 | 4995_focus_search_past_short |
| 4996 | dm - guys - - - | 20 | 4996_dm_guys__ |
| 4997 | article - help - home - swap - buy | 20 | 4997_article_help_home_swap |
| 4998 | transaction - note - old - - | 20 | 4998_transaction_note_old_ |
| 4999 | discord - enjoy - community - - | 20 | 4999_discord_enjoy_community_ |
| 5000 | potential - power - future - - | 20 | 5000_potential_power_future_ |
| 5001 | late - hour - action - today - | 20 | 5001_late_hour_action_today |
| 5002 | bot - ethereum - early - version - chain | 20 | 5002_bot_ethereum_early_version |
| 5003 | - - - - | 20 | 5003____ |
| 5004 | night - times - ready - good - morning | 20 | 5004_night_times_ready_good |
| 5005 | hand - option - form - announcement - mate | 20 | 5005_hand_option_form_announcement |
| 5006 | projects - ai - release - step - launch | 20 | 5006_projects_ai_release_step |
| 5007 | rules - case - global - links - private | 20 | 5007_rules_case_global_links |
| 5008 | volume - data - share - price - market | 20 | 5008_volume_data_share_price |
| 5009 | second - blockchain - bitcoin - volume - recent | 20 | 5009_second_blockchain_bitcoin_volume |
| 5010 | chance - special - previous - latest - worth | 20 | 5010_chance_special_previous_latest |
| 5011 | pool - liquidity - single - news - admin | 20 | 5011_pool_liquidity_single_news |
| 5012 | issue - build - difference - past - fine | 20 | 5012_issue_build_difference_past |
| 5013 | specific - crazy - left - latest - drop | 20 | 5013_specific_crazy_left_latest |
| 5014 | level - single - main - bit - fun | 20 | 5014_level_single_main_bit |
| 5015 | successful - free - time - crazy - old | 20 | 5015_successful_free_time_crazy |
| 5016 | super - biggest - interesting - period - lol | 20 | 5016_super_biggest_interesting_period |
| 5017 | report - unique - article - close - positive | 20 | 5017_report_unique_article_close |
| 5018 | damn - ways - word - building - chance | 20 | 5018_damn_ways_word_building |
| 5019 | dude - social - home - cause - wrong | 20 | 5019_dude_social_home_cause |
| 5020 | luck - unique - world - small - little | 20 | 5020_luck_unique_world_small |
| 5021 | price - luck - course - tomorrow - profit | 20 | 5021_price_luck_course_tomorrow |
| 5022 | latest - good - - - | 20 | 5022_latest_good__ |
| 5023 | rest - day - - - | 20 | 5023_rest_day__ |
| 5024 | thinking - luck - bad - cool - job | 20 | 5024_thinking_luck_bad_cool |
| 5025 | interested - phone - smart - money - home | 20 | 5025_interested_phone_smart_money |
| 5026 | thanks - work - good - channel - telegram | 20 | 5026_thanks_work_good_channel |
| 5027 | shit - word - dude - crazy - multiple | 20 | 5027_shit_word_dude_crazy |
| 5028 | - - - - | 20 | 5028____ |
| 5029 | recent - development - exchange - weeks - term | 20 | 5029_recent_development_exchange_weeks |
| 5030 | home - - - - | 20 | 5030_home___ |
| 5031 | event - change - low - high - trade | 20 | 5031_event_change_low_high |
| 5032 | research - spot - binance - guide - trading | 20 | 5032_research_spot_binance_guide |
| 5033 | higher - digital - trading - - | 20 | 5033_higher_digital_trading_ |
| 5034 | upcoming - announcement - listing - details - sign | 20 | 5034_upcoming_announcement_listing_details |
| 5035 | projects - market - - - | 20 | 5035_projects_market__ |
| 5036 | - - - - | 20 | 5036____ |
| 5037 | pool - change - reply - fair - users | 20 | 5037_pool_change_reply_fair |
| 5038 | past - value - year - nft - coins | 20 | 5038_past_value_year_nft |
| 5039 | transaction - btc - community - day - bitcoin | 20 | 5039_transaction_btc_community_day |
| 5040 | information - available - today - positive - site | 20 | 5040_information_available_today_positive |
| 5041 | personal - account - - - | 20 | 5041_personal_account__ |
| 5042 | total - simple - coins - bit - guide | 20 | 5042_total_simple_coins_bit |
| 5043 | attention - people - - - | 20 | 5043_attention_people__ |
| 5044 | btc - price - week - new - | 20 | 5044_btc_price_week_new |
| 5045 | ethereum - low - exchange - eth - volume | 20 | 5045_ethereum_low_exchange_eth |
| 5046 | type - protocol - contract - links - grow | 20 | 5046_type_protocol_contract_links |
| 5047 | feature - pool - liquidity - swap - holders | 20 | 5047_feature_pool_liquidity_swap |
| 5048 | ethereum - chain - smart - token - bitcoin | 20 | 5048_ethereum_chain_smart_token |
| 5049 | love - real - lol - today - | 20 | 5049_love_real_lol_today |
| 5050 | bitcoin - fees - real - - | 20 | 5050_bitcoin_fees_real_ |
| 5051 | bitcoin - - - - | 20 | 5051_bitcoin___ |
| 5052 | normal - total - profit - daily - code | 20 | 5052_normal_total_profit_daily |
| 5053 | questions - ama - minutes - space - utc | 20 | 5053_questions_ama_minutes_space |
| 5054 | email - head - form - contact - ama | 20 | 5054_email_head_form_contact |
| 5055 | coins - fees - transactions - blockchain - high | 20 | 5055_coins_fees_transactions_blockchain |
| 5056 | key - - - - | 20 | 5056_key___ |
| 5057 | app - games - daily - online - exciting | 20 | 5057_app_games_daily_online |
| 5058 | business - bitcoin - web - updates - | 20 | 5058_business_bitcoin_web_updates |
| 5059 | luck - happy - good - level - life | 20 | 5059_luck_happy_good_level |
| 5060 | private - words - kind - block - email | 20 | 5060_private_words_kind_block |
| 5061 | game - discord - issues - site - kind | 20 | 5061_game_discord_issues_site |
| 5062 | rate - ethereum - rewards - reward - network | 20 | 5062_rate_ethereum_rewards_reward |
| 5063 | pool - update - yesterday - high - rewards | 20 | 5063_pool_update_yesterday_high |
| 5064 | site - rewards - order - minimum - accounts | 20 | 5064_site_rewards_order_minimum |
| 5065 | article - guys - questions - work - free | 20 | 5065_article_guys_questions_work |
| 5066 | trading - - - - | 20 | 5066_trading___ |
| 5067 | rate - term - hard - search - protocol | 20 | 5067_rate_term_hard_search |
| 5068 | years - experience - - - | 20 | 5068_years_experience__ |
| 5069 | clear - turn - minutes - idea - mate | 20 | 5069_clear_turn_minutes_idea |
| 5070 | solution - positive - tho - trust - current | 20 | 5070_solution_positive_tho_trust |
| 5071 | latest - current - difference - release - stuff | 20 | 5071_latest_current_difference_release |
| 5072 | price - bitcoin - ready - - | 20 | 5072_price_bitcoin_ready_ |
| 5073 | awesome - glad - time - - | 20 | 5073_awesome_glad_time_ |
| 5074 | - - - - | 20 | 5074____ |
| 5075 | hour - ethereum - holders - volume - transactions | 20 | 5075_hour_ethereum_holders_volume |
| 5076 | new - transactions - network - wallet - | 20 | 5076_new_transactions_network_wallet |
| 5077 | holders - product - announcement - ethereum - defi | 20 | 5077_holders_product_announcement_ethereum |
| 5078 | transaction - note - investment - - | 20 | 5078_transaction_note_investment_ |
| 5079 | contact - dear - reason - list - account | 20 | 5079_contact_dear_reason_list |
| 5080 | - - - - | 20 | 5080____ |
| 5081 | - - - - | 20 | 5081____ |
| 5082 | problems - situation - ways - buddy - lots | 20 | 5082_problems_situation_ways_buddy |
| 5083 | deal - amazing - offer - new - free | 20 | 5083_deal_amazing_offer_new |
| 5084 | nft - amazing - rewards - enter - details | 20 | 5084_nft_amazing_rewards_enter |
| 5085 | shows - guy - country - lots - better | 20 | 5085_shows_guy_country_lots |
| 5086 | risk - funds - weeks - point - | 20 | 5086_risk_funds_weeks_point |
| 5087 | digital - weeks - week - report - total | 20 | 5087_digital_weeks_week_report |
| 5088 | point - tho - shit - thing - right | 20 | 5088_point_tho_shit_thing |
| 5089 | guide - bro - simple - matter - things | 20 | 5089_guide_bro_simple_matter |
| 5090 | family - past - profit - business - sir | 20 | 5090_family_past_profit_business |
| 5091 | near - future - service - new - users | 20 | 5091_near_future_service_new |
| 5092 | - - - - | 20 | 5092____ |
| 5093 | country - cryptocurrency - bitcoin - positive - | 20 | 5093_country_cryptocurrency_bitcoin_positive |
| 5094 | point - things - opinion - control - crazy | 20 | 5094_point_things_opinion_control |
| 5095 | bot - type - group - list - attention | 20 | 5095_bot_type_group_list |
| 5096 | cryptocurrency - - - - | 20 | 5096_cryptocurrency___ |
| 5097 | major - technology - blockchain - new - | 20 | 5097_major_technology_blockchain_new |
| 5098 | alot - chat - - - | 20 | 5098_alot_chat__ |
| 5099 | update - yesterday - second - old - different | 20 | 5099_update_yesterday_second_old |
| 5100 | previous - attention - kind - assets - data | 20 | 5100_previous_attention_kind_assets |
| 5101 | exchange - trade - crypto - message - kind | 20 | 5101_exchange_trade_crypto_message |
| 5102 | hour - quick - short - world - life | 20 | 5102_hour_quick_short_world |
| 5103 | report - crypto - rules - month - space | 20 | 5103_report_crypto_rules_month |
| 5104 | small - things - - - | 20 | 5104_small_things__ |
| 5105 | hand - dude - guess - comment - guy | 20 | 5105_hand_dude_guess_comment |
| 5106 | tokens - major - platform - country - home | 20 | 5106_tokens_major_platform_country |
| 5107 | guys - guy - good - - | 20 | 5107_guys_guy_good_ |
| 5108 | liquidity - btc - price - - | 20 | 5108_liquidity_btc_price_ |
| 5109 | normal - shit - bit - answer - person | 20 | 5109_normal_shit_bit_answer |
| 5110 | thing - thanks - right - platform - link | 20 | 5110_thing_thanks_right_platform |
| 5111 | nft - - - - | 20 | 5111_nft___ |
| 5112 | private - cryptocurrency - send - help - message | 20 | 5112_private_cryptocurrency_send_help |
| 5113 | news - - - - | 20 | 5113_news___ |
| 5114 | hours - new - days - - | 20 | 5114_hours_new_days_ |
| 5115 | happens - certain - fact - crazy - stuff | 20 | 5115_happens_certain_fact_crazy |
| 5116 | world - - - - | 20 | 5116_world___ |
| 5117 | major - tokens - reply - company - business | 20 | 5117_major_tokens_reply_company |
| 5118 | run - control - important - year - bit | 20 | 5118_run_control_important_year |
| 5119 | cool - words - kind - wrong - sorry | 20 | 5119_cool_words_kind_wrong |
| 5120 | night - ready - message - chat - world | 20 | 5120_night_ready_message_chat |
| 5121 | place - right - wrong - stuff - bro | 20 | 5121_place_right_wrong_stuff |
| 5122 | block - hour - ethereum - fast - biggest | 20 | 5122_block_hour_ethereum_fast |
| 5123 | admin - update - trading - word - trade | 20 | 5123_admin_update_trading_word |
| 5124 | release - key - public - service - trade | 20 | 5124_release_key_public_service |
| 5125 | event - fake - news - bitcoin - | 20 | 5125_event_fake_news_bitcoin |
| 5126 | help - situation - known - need - job | 20 | 5126_help_situation_known_need |
| 5127 | product - earn - - - | 20 | 5127_product_earn__ |
| 5128 | - - - - | 20 | 5128____ |
| 5129 | kind - thanks - nice - buddy - guy | 20 | 5129_kind_thanks_nice_buddy |
| 5130 | latest - week - blockchain - investors - network | 19 | 5130_latest_week_blockchain_investors |
| 5131 | night - today - thing - - | 19 | 5131_night_today_thing_ |
| 5132 | number - available - money - - | 19 | 5132_number_available_money_ |
| 5133 | half - okay - friend - month - good | 19 | 5133_half_okay_friend_month |
| 5134 | business - turn - assets - user - users | 19 | 5134_business_turn_assets_user |
| 5135 | assets - update - spot - exciting - development | 19 | 5135_assets_update_spot_exciting |
| 5136 | hours - minimum - deposit - project - withdrawal | 19 | 5136_hours_minimum_deposit_project |
| 5137 | glad - option - home - end - lol | 19 | 5137_glad_option_home_end |
| 5138 | rules - little - time - link - | 19 | 5138_rules_little_time_link |
| 5139 | event - protocol - rate - chart - months | 19 | 5139_event_protocol_rate_chart |
| 5140 | fee - service - trading - - | 19 | 5140_fee_service_trading_ |
| 5141 | known - ways - research - exchange - change | 19 | 5141_known_ways_research_exchange |
| 5142 | buddy - awesome - option - assets - reward | 19 | 5142_buddy_awesome_option_assets |
| 5143 | trade - trading - platform - link - | 19 | 5143_trade_trading_platform_link |
| 5144 | experience - problems - version - product - option | 19 | 5144_experience_problems_version_product |
| 5145 | active - space - - - | 19 | 5145_active_space__ |
| 5146 | family - hand - life - people - day | 19 | 5146_family_hand_life_people |
| 5147 | level - price - guys - potential - order | 19 | 5147_level_price_guys_potential |
| 5148 | welcome - hope - new - question - nice | 19 | 5148_welcome_hope_new_question |
| 5149 | moment - alot - fine - home - years | 19 | 5149_moment_alot_fine_home |
| 5150 | chat - simple - mate - big - | 19 | 5150_chat_simple_mate_big |
| 5151 | issues - works - project - - | 19 | 5151_issues_works_project_ |
| 5152 | - - - - | 19 | 5152____ |
| 5153 | specific - simple - home - public - technology | 19 | 5153_specific_simple_home_public |
| 5154 | opinion - known - comment - fair - word | 19 | 5154_opinion_known_comment_fair |
| 5155 | channel - telegram - today - link - amazing | 19 | 5155_channel_telegram_today_link |
| 5156 | web - active - version - nft - event | 19 | 5156_web_active_version_nft |
| 5157 | interesting - project - technology - hand - features | 19 | 5157_interesting_project_technology_hand |
| 5158 | possible - things - cause - doubt - | 19 | 5158_possible_things_cause_doubt |
| 5159 | tokens - digital - - - | 19 | 5159_tokens_digital__ |
| 5160 | country - stuff - guy - ones - simple | 19 | 5160_country_stuff_guy_ones |
| 5161 | bot - telegram - new - - | 19 | 5161_bot_telegram_new_ |
| 5162 | great - today - day - - | 19 | 5162_great_today_day_ |
| 5163 | yesterday - lots - possible - things - lot | 19 | 5163_yesterday_lots_possible_things |
| 5164 | ai - daily - article - sure - | 19 | 5164_ai_daily_article_sure |
| 5165 | end - btc - bitcoin - price - article | 19 | 5165_end_btc_bitcoin_price |
| 5166 | multiple - feature - tokens - token - research | 19 | 5166_multiple_feature_tokens_token |
| 5167 | development - business - friend - contact - similar | 19 | 5167_development_business_friend_contact |
| 5168 | crypto - certain - key - development - months | 19 | 5168_crypto_certain_key_development |
| 5169 | plans - bitcoin - report - trade - services | 19 | 5169_plans_bitcoin_report_trade |
| 5170 | chart - position - biggest - trade - buy | 19 | 5170_chart_position_biggest_trade |
| 5171 | link - big - chain - - | 19 | 5171_link_big_chain_ |
| 5172 | opportunity - awesome - successful - positive - bit | 19 | 5172_opportunity_awesome_successful_positive |
| 5173 | recent - interesting - guy - guide - years | 19 | 5173_recent_interesting_guy_guide |
| 5174 | deal - btc - nice - today - | 19 | 5174_deal_btc_nice_today |
| 5175 | rest - option - article - period - sense | 19 | 5175_rest_option_article_period |
| 5176 | latest - data - action - months - price | 19 | 5176_latest_data_action_months |
| 5177 | exchange - - - - | 19 | 5177_exchange___ |
| 5178 | claim - fee - sir - moment - sorry | 19 | 5178_claim_fee_sir_moment |
| 5179 | day - platform - amazing - projects - opportunity | 19 | 5179_day_platform_amazing_projects |
| 5180 | nice - day - great - fun - interesting | 19 | 5180_nice_day_great_fun |
| 5181 | case - public - - - | 19 | 5181_case_public__ |
| 5182 | service - news - - - | 19 | 5182_service_news__ |
| 5183 | safe - - - - | 19 | 5183_safe___ |
| 5184 | contact - sir - wish - team - friends | 19 | 5184_contact_sir_wish_team |
| 5185 | way - bitcoin - - - | 19 | 5185_way_bitcoin__ |
| 5186 | value - marketing - support - - | 19 | 5186_value_marketing_support_ |
| 5187 | previous - year - - - | 19 | 5187_previous_year__ |
| 5188 | dear - yesterday - friend - sure - | 19 | 5188_dear_yesterday_friend_sure |
| 5189 | cause - interesting - small - lost - man | 19 | 5189_cause_interesting_small_lost |
| 5190 | data - platform - - - | 19 | 5190_data_platform__ |
| 5191 | deposit - level - minimum - minutes - day | 19 | 5191_deposit_level_minimum_minutes |
| 5192 | dev - links - social - fair - launch | 19 | 5192_dev_links_social_fair |
| 5193 | point - new - day - - | 19 | 5193_point_new_day_ |
| 5194 | blockchain - comment - technology - video - | 19 | 5194_blockchain_comment_technology_video |
| 5195 | market - token - supply - hard - listing | 19 | 5195_market_token_supply_hard |
| 5196 | week - new - - - | 19 | 5196_week_new__ |
| 5197 | profits - day - free - days - | 19 | 5197_profits_day_free_days |
| 5198 | digital - head - guys - ready - group | 19 | 5198_digital_head_guys_ready |
| 5199 | close - reason - chance - sir - total | 19 | 5199_close_reason_chance_sir |
| 5200 | rewards - old - normal - supply - value | 19 | 5200_rewards_old_normal_supply |
| 5201 | low - account - price - rate - list | 19 | 5201_low_account_price_rate |
| 5202 | mate - news - group - great - | 19 | 5202_mate_news_group_great |
| 5203 | risk - control - research - trading - multiple | 19 | 5203_risk_control_research_trading |
| 5204 | certain - token - ready - available - wallet | 19 | 5204_certain_token_ready_available |
| 5205 | global - information - - - | 19 | 5205_global_information__ |
| 5206 | month - - - - | 19 | 5206_month___ |
| 5207 | assets - real - token - - | 19 | 5207_assets_real_token_ |
| 5208 | pls - fees - transaction - transactions - dude | 19 | 5208_pls_fees_transaction_transactions |
| 5209 | coins - bitcoin - - - | 19 | 5209_coins_bitcoin__ |
| 5210 | morning - week - good - great - family | 19 | 5210_morning_week_good_great |
| 5211 | main - data - - - | 19 | 5211_main_data__ |
| 5212 | main - rules - open - yesterday - stuff | 19 | 5212_main_rules_open_yesterday |
| 5213 | specific - stuff - guy - projects - sense | 19 | 5213_specific_stuff_guy_projects |
| 5214 | community - - - - | 19 | 5214_community___ |
| 5215 | video - business - specific - media - services | 19 | 5215_video_business_specific_media |
| 5216 | massive - launch - ama - supply - contract | 19 | 5216_massive_launch_ama_supply |
| 5217 | dude - face - word - wrong - little | 19 | 5217_dude_face_word_wrong |
| 5218 | public - opinion - app - minutes - higher | 19 | 5218_public_opinion_app_minutes |
| 5219 | hard - blockchain - block - hours - multiple | 19 | 5219_hard_blockchain_block_hours |
| 5220 | update - issue - thanks - sorry - updates | 19 | 5220_update_issue_thanks_sorry |
| 5221 | higher - - - - | 19 | 5221_higher___ |
| 5222 | active - free - - - | 19 | 5222_active_free__ |
| 5223 | position - chart - liquidity - buy - price | 19 | 5223_position_chart_liquidity_buy |
| 5224 | true - things - days - - | 19 | 5224_true_things_days_ |
| 5225 | sense - article - ones - option - period | 19 | 5225_sense_article_ones_option |
| 5226 | community - market - claim - site - limited | 19 | 5226_community_market_claim_site |
| 5227 | app - funds - available - wallet - | 19 | 5227_app_funds_available_wallet |
| 5228 | - - - - | 19 | 5228____ |
| 5229 | ai - - - - | 19 | 5229_ai___ |
| 5230 | site - simple - game - trust - second | 19 | 5230_site_simple_game_trust |
| 5231 | fee - claim - fees - high - details | 19 | 5231_fee_claim_fees_high |
| 5232 | hand - dude - fair - point - little | 19 | 5232_hand_dude_fair_point |
| 5233 | price - recent - selling - time - point | 19 | 5233_price_recent_selling_time |
| 5234 | night - positive - questions - twitter - latest | 19 | 5234_night_positive_questions_twitter |
| 5235 | action - liquidity - price - day - bitcoin | 19 | 5235_action_liquidity_price_day |
| 5236 | wallet - link - month - - | 19 | 5236_wallet_link_month_ |
| 5237 | cryptocurrency - new - - - | 19 | 5237_cryptocurrency_new__ |
| 5238 | chat - free - nice - great - | 19 | 5238_chat_free_nice_great |
| 5239 | share - - - - | 19 | 5239_share___ |
| 5240 | example - clear - perfect - real - sure | 19 | 5240_example_clear_perfect_real |
| 5241 | - - - - | 19 | 5241____ |
| 5242 | - - - - | 19 | 5242____ |
| 5243 | available - info - free - - | 19 | 5243_available_info_free_ |
| 5244 | data - btc - price - - | 19 | 5244_data_btc_price_ |
| 5245 | glad - games - article - game - early | 19 | 5245_glad_games_article_game |
| 5246 | mind - message - important - service - fake | 19 | 5246_mind_message_important_service |
| 5247 | clear - example - past - correct - issues | 19 | 5247_clear_example_past_correct |
| 5248 | site - free - profit - days - fee | 19 | 5248_site_free_profit_days |
| 5249 | crypto - opinion - article - rest - early | 19 | 5249_crypto_opinion_article_rest |
| 5250 | profit - trading - volume - option - hope | 19 | 5250_profit_trading_volume_option |
| 5251 | focus - - - - | 19 | 5251_focus___ |
| 5252 | tokens - date - site - code - unique | 19 | 5252_tokens_date_site_code |
| 5253 | course - update - guide - wallet - new | 19 | 5253_course_update_guide_wallet |
| 5254 | report - service - control - code - data | 19 | 5254_report_service_control_code |
| 5255 | twitter - crazy - attention - idea - position | 19 | 5255_twitter_crazy_attention_idea |
| 5256 | legit - company - - - | 19 | 5256_legit_company__ |
| 5257 | half - start - available - time - today | 19 | 5257_half_start_available_time |
| 5258 | admin - moment - case - reason - message | 19 | 5258_admin_moment_case_reason |
| 5259 | chat - lots - weeks - public - normal | 19 | 5259_chat_lots_weeks_public |
| 5260 | current - total - daily - - | 19 | 5260_current_total_daily_ |
| 5261 | reason - fake - things - easy - world | 19 | 5261_reason_fake_things_easy |
| 5262 | development - - - - | 19 | 5262_development___ |
| 5263 | assets - digital - investors - market - trading | 19 | 5263_assets_digital_investors_market |
| 5264 | place - guys - available - today - dm | 19 | 5264_place_guys_available_today |
| 5265 | deal - risk - liquidity - price - | 19 | 5265_deal_risk_liquidity_price |
| 5266 | step - type - features - case - public | 19 | 5266_step_type_features_case |
| 5267 | security - - - - | 19 | 5267_security___ |
| 5268 | wallet - - - - | 19 | 5268_wallet___ |
| 5269 | position - chart - holders - funds - buy | 19 | 5269_position_chart_holders_funds |
| 5270 | bitcoin - crypto - fact - home - address | 19 | 5270_bitcoin_crypto_fact_home |
| 5271 | games - fun - ready - message - | 19 | 5271_games_fun_ready_message |
| 5272 | problems - period - short - important - financial | 19 | 5272_problems_period_short_important |
| 5273 | links - utc - details - upcoming - excited | 19 | 5273_links_utc_details_upcoming |
| 5274 | chat - - - - | 19 | 5274_chat___ |
| 5275 | fair - game - special - country - main | 19 | 5275_fair_game_special_country |
| 5276 | investors - - - - | 19 | 5276_investors___ |
| 5277 | specific - opinion - idea - global - friend | 19 | 5277_specific_opinion_idea_global |
| 5278 | business - world - - - | 19 | 5278_business_world__ |
| 5279 | comment - guide - drop - reward - social | 19 | 5279_comment_guide_drop_reward |
| 5280 | time - - - - | 19 | 5280_time___ |
| 5281 | benefits - event - total - supply - video | 19 | 5281_benefits_event_total_supply |
| 5282 | plans - trade - news - digital - capital | 19 | 5282_plans_trade_news_digital |
| 5283 | fact - face - simple - man - issues | 19 | 5283_fact_face_simple_man |
| 5284 | specific - total - interested - previous - projects | 19 | 5284_specific_total_interested_previous |
| 5285 | bro - wrong - bad - right - day | 19 | 5285_bro_wrong_bad_right |
| 5286 | cool - web - idea - plan - different | 19 | 5286_cool_web_idea_plan |
| 5287 | damn - launch - community - case - crazy | 19 | 5287_damn_launch_community_case |
| 5288 | job - online - join - channel - telegram | 19 | 5288_job_online_join_channel |
| 5289 | fee - fees - kinda - correct - network | 19 | 5289_fee_fees_kinda_correct |
| 5290 | chat - life - group - telegram - link | 19 | 5290_chat_life_group_telegram |
| 5291 | nft - special - rewards - clear - form | 19 | 5291_nft_special_rewards_clear |
| 5292 | likely - weeks - country - hope - issues | 19 | 5292_likely_weeks_country_hope |
| 5293 | tokens - rate - crazy - years - times | 19 | 5293_tokens_rate_crazy_years |
| 5294 | number - public - shows - correct - sir | 19 | 5294_number_public_shows_correct |
| 5295 | guys - today - - - | 19 | 5295_guys_today__ |
| 5296 | control - past - week - - | 19 | 5296_control_past_week_ |
| 5297 | thing - email - sure - latest - ecosystem | 19 | 5297_thing_email_sure_latest |
| 5298 | interested - buying - risk - dm - deal | 19 | 5298_interested_buying_risk_dm |
| 5299 | term - long - benefits - - | 19 | 5299_term_long_benefits_ |
| 5300 | dude - specific - leave - stuff - attention | 19 | 5300_dude_specific_leave_stuff |
| 5301 | money - brother - situation - buddy - small | 19 | 5301_money_brother_situation_buddy |
| 5302 | control - data - users - single - personal | 19 | 5302_control_data_users_single |
| 5303 | chart - dev - contract - address - better | 19 | 5303_chart_dev_contract_address |
| 5304 | hard - close - correct - huge - country | 19 | 5304_hard_close_correct_huge |
| 5305 | binance - - - - | 19 | 5305_binance___ |
| 5306 | capital - - - - | 19 | 5306_capital___ |
| 5307 | half - action - reason - months - fee | 19 | 5307_half_action_reason_months |
| 5308 | worth - lol - bro - sure - work | 19 | 5308_worth_lol_bro_sure |
| 5309 | public - bitcoin - assets - news - today | 19 | 5309_public_bitcoin_assets_news |
| 5310 | technology - security - - - | 19 | 5310_technology_security__ |
| 5311 | normal - hand - dude - plan - experience | 19 | 5311_normal_hand_dude_plan |
| 5312 | volume - months - trading - tokens - market | 19 | 5312_volume_months_trading_tokens |
| 5313 | sorry - past - sure - possible - current | 19 | 5313_sorry_past_sure_possible |
| 5314 | week - step - day - - | 19 | 5314_week_step_day_ |
| 5315 | chain - pay - news - bitcoin - world | 19 | 5315_chain_pay_news_bitcoin |
| 5316 | old - access - life - - | 19 | 5316_old_access_life_ |
| 5317 | morning - night - good - - | 19 | 5317_morning_night_good_ |
| 5318 | bitcoin - real - btc - - | 19 | 5318_bitcoin_real_btc_ |
| 5319 | days - time - - - | 19 | 5319_days_time__ |
| 5320 | weeks - family - months - official - major | 19 | 5320_weeks_family_months_official |
| 5321 | public - ready - network - page - mate | 19 | 5321_public_ready_network_page |
| 5322 | nice - dear - fine - friend - course | 19 | 5322_nice_dear_fine_friend |
| 5323 | assets - account - - - | 19 | 5323_assets_account__ |
| 5324 | face - interested - okay - chat - nice | 19 | 5324_face_interested_okay_chat |
| 5325 | rewards - opportunity - information - sure - available | 19 | 5325_rewards_opportunity_information_sure |
| 5326 | tho - bit - course - case - morning | 19 | 5326_tho_bit_course_case |
| 5327 | fact - damn - point - difference - fair | 19 | 5327_fact_damn_point_difference |
| 5328 | dev - interesting - blockchain - position - situation | 19 | 5328_dev_interesting_blockchain_position |
| 5329 | access - online - total - site - rate | 19 | 5329_access_online_total_site |
| 5330 | announcement - list - information - benefits - contract | 19 | 5330_announcement_list_information_benefits |
| 5331 | minutes - job - business - sir - low | 19 | 5331_minutes_job_business_sir |
| 5332 | dm - - - - | 19 | 5332_dm___ |
| 5333 | liquidity - open - price - days - market | 19 | 5333_liquidity_open_price_days |
| 5334 | listing - technology - supply - research - minutes | 19 | 5334_listing_technology_supply_research |
| 5335 | ecosystem - massive - little - risk - level | 19 | 5335_ecosystem_massive_little_risk |
| 5336 | close - guy - financial - space - man | 19 | 5336_close_guy_financial_space |
| 5337 | lot - - - - | 19 | 5337_lot___ |
| 5338 | article - world - way - idea - term | 19 | 5338_article_world_way_idea |
| 5339 | benefits - - - - | 19 | 5339_benefits___ |
| 5340 | assets - company - - - | 19 | 5340_assets_company__ |
| 5341 | sense - example - form - thing - head | 19 | 5341_sense_example_form_thing |
| 5342 | rate - month - - - | 19 | 5342_rate_month__ |
| 5343 | - - - - | 19 | 5343____ |
| 5344 | ama - specific - problems - update - latest | 19 | 5344_ama_specific_problems_update |
| 5345 | space - protocol - features - similar - head | 19 | 5345_space_protocol_features_similar |
| 5346 | accounts - trust - wallet - business - team | 19 | 5346_accounts_trust_wallet_business |
| 5347 | ai - world - platform - company - | 19 | 5347_ai_world_platform_company |
| 5348 | - - - - | 19 | 5348____ |
| 5349 | rules - change - - - | 19 | 5349_rules_change__ |
| 5350 | - - - - | 19 | 5350____ |
| 5351 | normal - ok - later - bit - date | 19 | 5351_normal_ok_later_bit |
| 5352 | situation - positive - supply - case - action | 19 | 5352_situation_positive_supply_case |
| 5353 | check - - - - | 19 | 5353_check___ |
| 5354 | market - - - - | 19 | 5354_market___ |
| 5355 | job - fake - cryptocurrency - coins - tokens | 19 | 5355_job_fake_cryptocurrency_coins |
| 5356 | data - control - example - network - case | 19 | 5356_data_control_example_network |
| 5357 | normal - close - rest - games - code | 19 | 5357_normal_close_rest_games |
| 5358 | current - case - successful - - | 19 | 5358_current_case_successful_ |
| 5359 | likely - major - rest - early - country | 19 | 5359_likely_major_rest_early |
| 5360 | - - - - | 19 | 5360____ |
| 5361 | report - - - - | 19 | 5361_report___ |
| 5362 | question - cool - form - second - interested | 19 | 5362_question_cool_form_second |
| 5363 | - - - - | 19 | 5363____ |
| 5364 | - - - - | 19 | 5364____ |
| 5365 | moment - bit - left - amazing - things | 19 | 5365_moment_bit_left_amazing |
| 5366 | code - assets - rewards - free - app | 19 | 5366_code_assets_rewards_free |
| 5367 | accounts - funds - twitter - website - worth | 19 | 5367_accounts_funds_twitter_website |
| 5368 | trade - link - - - | 19 | 5368_trade_link__ |
| 5369 | job - daily - work - person - real | 19 | 5369_job_daily_work_person |
| 5370 | life - - - - | 19 | 5370_life___ |
| 5371 | deposit - financial - cryptocurrency - action - details | 19 | 5371_deposit_financial_cryptocurrency_action |
| 5372 | data - couple - feature - super - space | 19 | 5372_data_couple_feature_super |
| 5373 | clear - multiple - protocol - transactions - users | 19 | 5373_clear_multiple_protocol_transactions |
| 5374 | country - benefits - project - lot - couple | 19 | 5374_country_benefits_project_lot |
| 5375 | cryptocurrency - team - best - better - project | 19 | 5375_cryptocurrency_team_best_better |
| 5376 | date - near - specific - launch - fact | 19 | 5376_date_near_specific_launch |
| 5377 | security - smart - experience - transaction - chain | 19 | 5377_security_smart_experience_transaction |
| 5378 | site - term - deposit - daily - simple | 19 | 5378_site_term_deposit_daily |
| 5379 | legit - site - quick - bad - fake | 19 | 5379_legit_site_quick_bad |
| 5380 | email - community - - - | 19 | 5380_email_community__ |
| 5381 | exchange - word - way - couple - current | 19 | 5381_exchange_word_way_couple |
| 5382 | best - unique - number - - | 19 | 5382_best_unique_number_ |
| 5383 | multiple - features - feature - quick - smart | 19 | 5383_multiple_features_feature_quick |
| 5384 | protocol - complete - product - world - problems | 19 | 5384_protocol_complete_product_world |
| 5385 | ai - public - technology - personal - potential | 19 | 5385_ai_public_technology_personal |
| 5386 | night - reason - day - - | 19 | 5386_night_reason_day_ |
| 5387 | wrong - months - real - things - | 19 | 5387_wrong_months_real_things |
| 5388 | opportunity - biggest - ethereum - use - platform | 19 | 5388_opportunity_biggest_ethereum_use |
| 5389 | rules - unique - nft - worth - form | 19 | 5389_rules_unique_nft_worth |
| 5390 | article - research - half - big - times | 19 | 5390_article_research_half_big |
| 5391 | - - - - | 19 | 5391____ |
| 5392 | man - buddy - nice - fine - good | 19 | 5392_man_buddy_nice_fine |
| 5393 | ethereum - fee - safe - trade - | 19 | 5393_ethereum_fee_safe_trade |
| 5394 | group - rules - sense - meet - course | 19 | 5394_group_rules_sense_meet |
| 5395 | protocol - ethereum - network - technology - ecosystem | 19 | 5395_protocol_ethereum_network_technology |
| 5396 | technology - site - course - channel - video | 19 | 5396_technology_site_course_channel |
| 5397 | risk - platform - users - - | 19 | 5397_risk_platform_users_ |
| 5398 | luck - updates - claim - wallet - tokens | 19 | 5398_luck_updates_claim_wallet |
| 5399 | twitter - news - massive - biggest - technology | 18 | 5399_twitter_news_massive_biggest |
| 5400 | binance - certain - friends - rewards - note | 18 | 5400_binance_certain_friends_rewards |
| 5401 | important - thing - people - things - life | 18 | 5401_important_thing_people_things |
| 5402 | report - - - - | 18 | 5402_report___ |
| 5403 | issue - thank - way - - | 18 | 5403_issue_thank_way_ |
| 5404 | - - - - | 18 | 5404____ |
| 5405 | couple - word - idea - rate - little | 18 | 5405_couple_word_idea_rate |
| 5406 | rest - shit - reason - risk - week | 18 | 5406_rest_shit_reason_risk |
| 5407 | links - free - payment - - | 18 | 5407_links_free_payment_ |
| 5408 | - - - - | 18 | 5408____ |
| 5409 | matter - situation - sure - time - development | 18 | 5409_matter_situation_sure_time |
| 5410 | major - nice - lol - - | 18 | 5410_major_nice_lol_ |
| 5411 | word - multiple - higher - question - little | 18 | 5411_word_multiple_higher_question |
| 5412 | fine - alright - ways - buddy - bit | 18 | 5412_fine_alright_ways_buddy |
| 5413 | assets - digital - - - | 18 | 5413_assets_digital__ |
| 5414 | assets - buddy - option - minutes - guide | 18 | 5414_assets_buddy_option_minutes |
| 5415 | block - security - transaction - rewards - network | 18 | 5415_block_security_transaction_rewards |
| 5416 | glad - right - - - | 18 | 5416_glad_right__ |
| 5417 | trade - payment - link - correct - hard | 18 | 5417_trade_payment_link_correct |
| 5418 | spot - thinking - cool - position - short | 18 | 5418_spot_thinking_cool_position |
| 5419 | development - mate - block - interested - lol | 18 | 5419_development_mate_block_interested |
| 5420 | earlier - report - crypto - public - | 18 | 5420_earlier_report_crypto_public |
| 5421 | issues - works - project - - | 18 | 5421_issues_works_project_ |
| 5422 | - - - - | 18 | 5422____ |
| 5423 | change - example - global - recent - kind | 18 | 5423_change_example_global_recent |
| 5424 | transactions - weeks - bitcoin - late - hour | 18 | 5424_transactions_weeks_bitcoin_late |
| 5425 | hours - real - available - free - | 18 | 5425_hours_real_available_free |
| 5426 | spot - assets - users - withdraw - trading | 18 | 5426_spot_assets_users_withdraw |
| 5427 | high - - - - | 18 | 5427_high___ |
| 5428 | hours - new - - - | 18 | 5428_hours_new__ |
| 5429 | games - chart - holders - funds - website | 18 | 5429_games_chart_holders_funds |
| 5430 | pay - investment - - - | 18 | 5430_pay_investment__ |
| 5431 | certain - payment - new - - | 18 | 5431_certain_payment_new_ |
| 5432 | visit - ready - launch - world - chain | 18 | 5432_visit_ready_launch_world |
| 5433 | withdrawal - rate - transaction - working - days | 18 | 5433_withdrawal_rate_transaction_working |
| 5434 | opinion - interesting - couple - strong - token | 18 | 5434_opinion_interesting_couple_strong |
| 5435 | perfect - world - guy - - | 18 | 5435_perfect_world_guy_ |
| 5436 | bitcoin - article - technology - important - private | 18 | 5436_bitcoin_article_technology_important |
| 5437 | nft - key - holders - swap - trade | 18 | 5437_nft_key_holders_swap |
| 5438 | rules - crypto - - - | 18 | 5438_rules_crypto__ |
| 5439 | current - - - - | 18 | 5439_current___ |
| 5440 | trading - wallet - link - - | 18 | 5440_trading_wallet_link_ |
| 5441 | country - case - family - hard - projects | 18 | 5441_country_case_family_hard |
| 5442 | early - minimum - site - daily - complete | 18 | 5442_early_minimum_site_daily |
| 5443 | yesterday - games - home - profit - today | 18 | 5443_yesterday_games_home_profit |
| 5444 | little - shit - wallet - time - | 18 | 5444_little_shit_wallet_time |
| 5445 | bad - platform - fact - business - recent | 18 | 5445_bad_platform_fact_business |
| 5446 | friend - single - current - dear - business | 18 | 5446_friend_single_current_dear |
| 5447 | major - wrong - point - small - true | 18 | 5447_major_wrong_point_small |
| 5448 | amazing - capital - happy - thanks - life | 18 | 5448_amazing_capital_happy_thanks |
| 5449 | website - telegram - link - - | 18 | 5449_website_telegram_link_ |
| 5450 | feature - special - cool - updates - users | 18 | 5450_feature_special_cool_updates |
| 5451 | data - price - - - | 18 | 5451_data_price__ |
| 5452 | awesome - amazing - links - open - fast | 18 | 5452_awesome_amazing_links_open |
| 5453 | announcement - news - official - updates - feature | 18 | 5453_announcement_news_official_updates |
| 5454 | sir - channel - job - help - telegram | 18 | 5454_sir_channel_job_help |
| 5455 | rewards - days - update - app - second | 18 | 5455_rewards_days_update_app |
| 5456 | project - luck - major - perfect - success | 18 | 5456_project_luck_major_perfect |
| 5457 | projects - main - cool - rate - thing | 18 | 5457_projects_main_cool_rate |
| 5458 | ai - - - - | 18 | 5458_ai___ |
| 5459 | single - user - total - fees - trade | 18 | 5459_single_user_total_fees |
| 5460 | ama - business - place - live - binance | 18 | 5460_ama_business_place_live |
| 5461 | open - interested - awesome - friend - sir | 18 | 5461_open_interested_awesome_friend |
| 5462 | years - week - wallet - - | 18 | 5462_years_week_wallet_ |
| 5463 | user - important - true - key - public | 18 | 5463_user_important_true_key |
| 5464 | yesterday - future - today - tomorrow - | 18 | 5464_yesterday_future_today_tomorrow |
| 5465 | current - early - months - value - moment | 18 | 5465_current_early_months_value |
| 5466 | games - specific - playing - article - tho | 18 | 5466_games_specific_playing_article |
| 5467 | step - opinion - small - rest - months | 18 | 5467_step_opinion_small_rest |
| 5468 | positive - idea - holders - number - months | 18 | 5468_positive_idea_holders_number |
| 5469 | action - total - value - month - price | 18 | 5469_action_total_value_month |
| 5470 | ethereum - dude - swap - difference - better | 18 | 5470_ethereum_dude_swap_difference |
| 5471 | volume - change - update - supply - hours | 18 | 5471_volume_change_update_supply |
| 5472 | claim - latest - users - - | 18 | 5472_claim_latest_users_ |
| 5473 | attention - links - official - - | 18 | 5473_attention_links_official_ |
| 5474 | ai - past - ones - rest - matter | 18 | 5474_ai_past_ones_rest |
| 5475 | personal - account - guys - - | 18 | 5475_personal_account_guys_ |
| 5476 | details - great - - - | 18 | 5476_details_great__ |
| 5477 | binance - article - word - second - updates | 18 | 5477_binance_article_word_second |
| 5478 | main - binance - brother - use - issue | 18 | 5478_main_binance_brother_use |
| 5479 | btc - payment - account - group - free | 18 | 5479_btc_payment_account_group |
| 5480 | luck - simple - ready - good - bad | 18 | 5480_luck_simple_ready_good |
| 5481 | personal - private - lots - normal - product | 18 | 5481_personal_private_lots_normal |
| 5482 | private - users - real - - | 18 | 5482_private_users_real_ |
| 5483 | way - - - - | 18 | 5483_way___ |
| 5484 | future - - - - | 18 | 5484_future___ |
| 5485 | price - - - - | 18 | 5485_price___ |
| 5486 | guys - today - happy - day - new | 18 | 5486_guys_today_happy_day |
| 5487 | worth - bitcoin - - - | 18 | 5487_worth_bitcoin__ |
| 5488 | strong - man - easy - life - ppl | 18 | 5488_strong_man_easy_life |
| 5489 | report - private - admin - links - fake | 18 | 5489_report_private_admin_links |
| 5490 | list - shit - site - bit - case | 18 | 5490_list_shit_site_bit |
| 5491 | rate - place - market - - | 18 | 5491_rate_place_market_ |
| 5492 | announcement - details - supply - total - sure | 18 | 5492_announcement_details_supply_total |
| 5493 | service - video - available - information - lol | 18 | 5493_service_video_available_information |
| 5494 | claim - main - total - account - wallet | 18 | 5494_claim_main_total_account |
| 5495 | dev - coins - projects - future - coin | 18 | 5495_dev_coins_projects_future |
| 5496 | success - - - - | 18 | 5496_success___ |
| 5497 | word - awesome - glad - post - words | 18 | 5497_word_awesome_glad_post |
| 5498 | data - - - - | 18 | 5498_data___ |
| 5499 | connect - trust - step - wallet - write | 18 | 5499_connect_trust_step_wallet |
| 5500 | control - weeks - true - country - things | 18 | 5500_control_weeks_true_country |
| 5501 | building - team - look - old - product | 18 | 5501_building_team_look_old |
| 5502 | option - words - fee - services - withdrawal | 18 | 5502_option_words_fee_services |
| 5503 | app - official - website - group - daily | 18 | 5503_app_official_website_group |
| 5504 | days - trading - crypto - past - stuff | 18 | 5504_days_trading_crypto_past |
| 5505 | dude - - - - | 18 | 5505_dude___ |
| 5506 | project - - - - | 18 | 5506_project___ |
| 5507 | single - message - chat - talk - type | 18 | 5507_single_message_chat_talk |
| 5508 | true - luck - friends - nice - good | 18 | 5508_true_luck_friends_nice |
| 5509 | massive - benefits - ama - talk - listing | 18 | 5509_massive_benefits_ama_talk |
| 5510 | short - fee - ethereum - safe - | 18 | 5510_short_fee_ethereum_safe |
| 5511 | risk - blockchain - code - services - network | 18 | 5511_risk_blockchain_code_services |
| 5512 | stuff - games - game - number - year | 18 | 5512_stuff_games_game_number |
| 5513 | form - different - project - investment - | 18 | 5513_form_different_project_investment |
| 5514 | - - - - | 18 | 5514____ |
| 5515 | super - interesting - cryptocurrency - project - future | 18 | 5515_super_interesting_cryptocurrency_project |
| 5516 | digital - blockchain - - - | 18 | 5516_digital_blockchain__ |
| 5517 | bot - number - yesterday - stuff - ones | 18 | 5517_bot_number_yesterday_stuff |
| 5518 | project - alot - interested - good - | 18 | 5518_project_alot_interested_good |
| 5519 | ethereum - rewards - week - rate - increase | 18 | 5519_ethereum_rewards_week_rate |
| 5520 | clear - opportunity - guys - brother - focus | 18 | 5520_clear_opportunity_guys_brother |
| 5521 | defi - close - spot - words - home | 18 | 5521_defi_close_spot_words |
| 5522 | visit - private - claim - live - exchange | 18 | 5522_visit_private_claim_live |
| 5523 | public - - - - | 18 | 5523_public___ |
| 5524 | amazing - capital - happy - thanks - life | 18 | 5524_amazing_capital_happy_thanks |
| 5525 | trading - bitcoin - - - | 18 | 5525_trading_bitcoin__ |
| 5526 | awesome - work - great - guy - site | 18 | 5526_awesome_work_great_guy |
| 5527 | available - info - time - - | 18 | 5527_available_info_time_ |
| 5528 | thanks - sounds - luck - amazing - sure | 18 | 5528_thanks_sounds_luck_amazing |
| 5529 | telegram - current - times - launch - public | 18 | 5529_telegram_current_times_launch |
| 5530 | job - country - bit - quick - mate | 18 | 5530_job_country_bit_quick |
| 5531 | position - plan - liquidity - market - positive | 18 | 5531_position_plan_liquidity_market |
| 5532 | site - deal - holders - guys - ready | 18 | 5532_site_deal_holders_guys |
| 5533 | assets - security - media - social - users | 18 | 5533_assets_security_media_social |
| 5534 | global - - - - | 18 | 5534_global___ |
| 5535 | chat - people - holders - online - community | 18 | 5535_chat_people_holders_online |
| 5536 | past - ethereum - total - eth - previous | 18 | 5536_past_ethereum_total_eth |
| 5537 | difference - nice - point - mate - long | 18 | 5537_difference_nice_point_mate |
| 5538 | article - correct - supply - data - information | 18 | 5538_article_correct_supply_data |
| 5539 | moment - time - bit - minutes - sure | 18 | 5539_moment_time_bit_minutes |
| 5540 | course - mate - important - situation - experience | 18 | 5540_course_mate_important_situation |
| 5541 | hour - volume - holders - transactions - price | 18 | 5541_hour_volume_holders_transactions |
| 5542 | sense - feature - known - wallet - interesting | 18 | 5542_sense_feature_known_wallet |
| 5543 | issue - days - positive - lots - ecosystem | 18 | 5543_issue_days_positive_lots |
| 5544 | interested - secure - - - | 18 | 5544_interested_secure__ |
| 5545 | assets - projects - real - world - | 18 | 5545_assets_projects_real_world |
| 5546 | reward - free - worth - user - minimum | 18 | 5546_reward_free_worth_user |
| 5547 | night - - - - | 18 | 5547_night___ |
| 5548 | channel - telegram - link - - | 18 | 5548_channel_telegram_link_ |
| 5549 | rules - - - - | 18 | 5549_rules___ |
| 5550 | home - work - daily - people - current | 18 | 5550_home_work_daily_people |
| 5551 | thing - situation - luck - point - times | 18 | 5551_thing_situation_luck_point |
| 5552 | web - similar - product - previous - ones | 18 | 5552_web_similar_product_previous |
| 5553 | listing - coin - clear - investors - btc | 18 | 5553_listing_coin_clear_investors |
| 5554 | past - total - days - market - current | 18 | 5554_past_total_days_market |
| 5555 | worth - minimum - address - users - event | 18 | 5555_worth_minimum_address_users |
| 5556 | holders - future - happy - man - nice | 18 | 5556_holders_future_happy_man |
| 5557 | buddy - bad - single - thing - fake | 18 | 5557_buddy_bad_single_thing |
| 5558 | visit - announcement - details - total - supply | 18 | 5558_visit_announcement_details_total |
| 5559 | contract - address - note - safe - games | 18 | 5559_contract_address_note_safe |
| 5560 | supply - year - special - claim - win | 18 | 5560_supply_year_special_claim |
| 5561 | event - fee - trading - - | 18 | 5561_event_fee_trading_ |
| 5562 | rules - chance - worth - minimum - active | 18 | 5562_rules_chance_worth_minimum |
| 5563 | shit - bit - - - | 18 | 5563_shit_bit__ |
| 5564 | users - account - - - | 18 | 5564_users_account__ |
| 5565 | fast - today - withdrawal - platform - | 18 | 5565_fast_today_withdrawal_platform |
| 5566 | better - right - wrong - time - early | 18 | 5566_better_right_wrong_time |
| 5567 | visit - order - event - stuff - job | 18 | 5567_visit_order_event_stuff |
| 5568 | care - drop - pm - upcoming - digital | 18 | 5568_care_drop_pm_upcoming |
| 5569 | launch - global - number - marketing - huge | 18 | 5569_launch_global_number_marketing |
| 5570 | current - future - price - available - | 18 | 5570_current_future_price_available |
| 5571 | deal - drop - massive - rate - real | 18 | 5571_deal_drop_massive_rate |
| 5572 | - - - - | 18 | 5572____ |
| 5573 | sense - positive - massive - kind - answer | 18 | 5573_sense_positive_massive_kind |
| 5574 | user - experience - interesting - features - users | 18 | 5574_user_experience_interesting_features |
| 5575 | search - drop - daily - media - social | 18 | 5575_search_drop_daily_media |
| 5576 | sorry - legit - man - guys - post | 18 | 5576_sorry_legit_man_guys |
| 5577 | digital - technology - blockchain - action - assets | 18 | 5577_digital_technology_blockchain_action |
| 5578 | public - marketing - rewards - clear - rules | 18 | 5578_public_marketing_rewards_clear |
| 5579 | period - dude - shit - night - cool | 18 | 5579_period_dude_shit_night |
| 5580 | trading - couple - late - luck - shit | 18 | 5580_trading_couple_late_luck |
| 5581 | defi - - - - | 18 | 5581_defi___ |
| 5582 | means - payment - - - | 18 | 5582_means_payment__ |
| 5583 | official - crypto - new - - | 18 | 5583_official_crypto_new_ |
| 5584 | latest - updates - links - discord - list | 18 | 5584_latest_updates_links_discord |
| 5585 | - - - - | 18 | 5585____ |
| 5586 | late - months - research - profits - secure | 18 | 5586_late_months_research_profits |
| 5587 | previous - normal - quick - liquidity - value | 18 | 5587_previous_normal_quick_liquidity |
| 5588 | kind - wonder - crazy - nice - person | 18 | 5588_kind_wonder_crazy_nice |
| 5589 | country - lol - fun - reason - talk | 18 | 5589_country_lol_fun_reason |
| 5590 | claim - information - - - | 18 | 5590_claim_information__ |
| 5591 | - - - - | 18 | 5591____ |
| 5592 | today - good - - - | 18 | 5592_today_good__ |
| 5593 | risk - market - fair - current - low | 18 | 5593_risk_market_fair_current |
| 5594 | massive - country - run - point - holders | 18 | 5594_massive_country_run_point |
| 5595 | minimum - claim - biggest - service - payment | 18 | 5595_minimum_claim_biggest_service |
| 5596 | value - - - - | 18 | 5596_value___ |
| 5597 | time - - - - | 18 | 5597_time___ |
| 5598 | liquidity - low - new - higher - term | 18 | 5598_liquidity_low_new_higher |
| 5599 | video - thanks - daily - eth - people | 18 | 5599_video_thanks_daily_eth |
| 5600 | great - week - - - | 18 | 5600_great_week__ |
| 5601 | supply - value - business - user - | 18 | 5601_supply_value_business_user |
| 5602 | deposit - main - minimum - level - note | 18 | 5602_deposit_main_minimum_level |
| 5603 | people - start - kind - morning - nice | 18 | 5603_people_start_kind_morning |
| 5604 | date - news - channel - official - announcement | 18 | 5604_date_news_channel_official |
| 5605 | country - massive - benefits - contact - telegram | 18 | 5605_country_massive_benefits_contact |
| 5606 | cool - specific - problems - major - late | 18 | 5606_cool_specific_problems_major |
| 5607 | weeks - yesterday - months - shit - night | 18 | 5607_weeks_yesterday_months_shit |
| 5608 | early - late - face - increase - success | 18 | 5608_early_late_face_increase |
| 5609 | difference - family - members - guys - life | 18 | 5609_difference_family_members_guys |
| 5610 | bitcoin - new - investment - - | 18 | 5610_bitcoin_new_investment_ |
| 5611 | questions - things - blockchain - users - people | 18 | 5611_questions_things_blockchain_users |
| 5612 | investment - - - - | 18 | 5612_investment___ |
| 5613 | development - digital - community - trust - idea | 18 | 5613_development_digital_community_trust |
| 5614 | similar - information - shit - lots - run | 18 | 5614_similar_information_shit_lots |
| 5615 | trade - - - - | 18 | 5615_trade___ |
| 5616 | building - focus - social - members - services | 18 | 5616_building_focus_social_members |
| 5617 | buying - year - opportunity - - | 18 | 5617_buying_year_opportunity_ |
| 5618 | strong - major - play - years - sure | 18 | 5618_strong_major_play_years |
| 5619 | problem - bitcoin - rules - course - chance | 18 | 5619_problem_bitcoin_rules_course |
| 5620 | luck - minutes - cool - times - point | 18 | 5620_luck_minutes_cool_times |
| 5621 | open - bitcoin - person - development - action | 18 | 5621_open_bitcoin_person_development |
| 5622 | people - - - - | 18 | 5622_people___ |
| 5623 | holders - rewards - - - | 18 | 5623_holders_rewards__ |
| 5624 | protocol - data - known - secure - security | 18 | 5624_protocol_data_known_secure |
| 5625 | extra - page - payment - great - best | 18 | 5625_extra_page_payment_great |
| 5626 | things - fact - word - rest - cool | 18 | 5626_things_fact_word_rest |
| 5627 | months - trade - opinion - link - month | 18 | 5627_months_trade_opinion_link |
| 5628 | available - note - rate - payment - offer | 18 | 5628_available_note_rate_payment |
| 5629 | ca - search - related - example - process | 18 | 5629_ca_search_related_example |
| 5630 | - - - - | 18 | 5630____ |
| 5631 | value - assets - user - details - share | 18 | 5631_value_assets_user_details |
| 5632 | date - dude - past - year - lol | 18 | 5632_date_dude_past_year |
| 5633 | technology - digital - blockchain - - | 18 | 5633_technology_digital_blockchain_ |
| 5634 | worth - buying - single - user - total | 18 | 5634_worth_buying_single_user |
| 5635 | guess - game - event - date - reward | 18 | 5635_guess_game_event_date |
| 5636 | cool - bitcoin - point - power - position | 18 | 5636_cool_bitcoin_point_power |
| 5637 | binance - user - exchange - code - free | 18 | 5637_binance_user_exchange_code |
| 5638 | research - important - cryptocurrency - market - true | 18 | 5638_research_important_cryptocurrency_market |
| 5639 | holders - recent - yesterday - update - case | 18 | 5639_holders_recent_yesterday_update |
| 5640 | excited - home - care - family - thanks | 18 | 5640_excited_home_care_family |
| 5641 | early - holders - launch - private - reward | 18 | 5641_early_holders_launch_private |
| 5642 | home - welcome - channel - business - contact | 18 | 5642_home_welcome_channel_business |
| 5643 | step - technology - example - number - attention | 18 | 5643_step_technology_example_number |
| 5644 | deal - btc - - - | 18 | 5644_deal_btc__ |
| 5645 | late - single - today - platform - investment | 18 | 5645_late_single_today_platform |
| 5646 | comment - updates - major - search - report | 18 | 5646_comment_updates_major_search |
| 5647 | coin - legit - coins - fact - official | 18 | 5647_coin_legit_coins_fact |
| 5648 | morning - little - night - cool - position | 18 | 5648_morning_little_night_cool |
| 5649 | previous - holders - huge - media - power | 18 | 5649_previous_holders_huge_media |
| 5650 | fact - watch - buy - - | 18 | 5650_fact_watch_buy_ |
| 5651 | version - question - shit - latest - quick | 18 | 5651_version_question_shit_latest |
| 5652 | trade - link - - - | 18 | 5652_trade_link__ |
| 5653 | similar - major - glad - interesting - release | 18 | 5653_similar_major_glad_interesting |
| 5654 | type - higher - short - number - multiple | 18 | 5654_type_higher_short_number |
| 5655 | current - social - media - projects - specific | 18 | 5655_current_social_media_projects |
| 5656 | words - new - - - | 18 | 5656_words_new__ |
| 5657 | legit - end - extra - online - way | 18 | 5657_legit_end_extra_online |
| 5658 | person - yesterday - bad - lot - ok | 18 | 5658_person_yesterday_bad_lot |
| 5659 | alot - single - future - time - | 18 | 5659_alot_single_future_time |
| 5660 | times - - - - | 18 | 5660_times___ |
| 5661 | quick - answer - space - question - perfect | 18 | 5661_quick_answer_space_question |
| 5662 | form - legit - cryptocurrency - fees - extra | 18 | 5662_form_legit_cryptocurrency_fees |
| 5663 | type - build - updates - easy - way | 18 | 5663_type_build_updates_easy |
| 5664 | world - day - - - | 18 | 5664_world_day__ |
| 5665 | bot - problems - ways - earn - word | 18 | 5665_bot_problems_ways_earn |
| 5666 | smart - contract - network - protocol - data | 18 | 5666_smart_contract_network_protocol |
| 5667 | chain - smart - binance - near - guide | 18 | 5667_chain_smart_binance_near |
| 5668 | sorry - brother - special - situation - normal | 18 | 5668_sorry_brother_special_situation |
| 5669 | holders - buddy - man - chance - future | 18 | 5669_holders_buddy_man_chance |
| 5670 | - - - - | 18 | 5670____ |
| 5671 | unique - business - blockchain - - | 18 | 5671_unique_business_blockchain_ |
| 5672 | person - ama - ecosystem - marketing - questions | 18 | 5672_person_ama_ecosystem_marketing |
| 5673 | friend - deposit - members - profit - group | 18 | 5673_friend_deposit_members_profit |
| 5674 | data - - - - | 18 | 5674_data___ |
| 5675 | previous - order - - - | 18 | 5675_previous_order__ |
| 5676 | user - - - - | 18 | 5676_user___ |
| 5677 | weeks - ready - lot - sure - capital | 18 | 5677_weeks_ready_lot_sure |
| 5678 | - - - - | 18 | 5678____ |
| 5679 | questions - attention - matter - answer - sorry | 17 | 5679_questions_attention_matter_answer |
| 5680 | half - business - buying - form - kind | 17 | 5680_half_business_buying_form |
| 5681 | - - - - | 17 | 5681____ |
| 5682 | exchange - address - wallet - transaction - personal | 17 | 5682_exchange_address_wallet_transaction |
| 5683 | ecosystem - biggest - product - important - technology | 17 | 5683_ecosystem_biggest_product_important |
| 5684 | period - hour - bit - point - small | 17 | 5684_period_hour_bit_point |
| 5685 | page - active - fast - channel - available | 17 | 5685_page_active_fast_channel |
| 5686 | technology - global - report - financial - security | 17 | 5686_technology_global_report_financial |
| 5687 | coins - coin - ways - project - projects | 17 | 5687_coins_coin_ways_project |
| 5688 | legit - active - payment - company - investment | 17 | 5688_legit_active_payment_company |
| 5689 | data - community - specific - ai - public | 17 | 5689_data_community_specific_ai |
| 5690 | rules - - - - | 17 | 5690_rules___ |
| 5691 | research - man - - - | 17 | 5691_research_man__ |
| 5692 | pool - participate - deal - details - event | 17 | 5692_pool_participate_deal_details |
| 5693 | main - specific - fact - moment - cool | 17 | 5693_main_specific_fact_moment |
| 5694 | good - fine - great - - | 17 | 5694_good_fine_great_ |
| 5695 | luck - period - pool - complete - buy | 17 | 5695_luck_period_pool_complete |
| 5696 | alright - certain - main - case - friend | 17 | 5696_alright_certain_main_case |
| 5697 | - - - - | 17 | 5697____ |
| 5698 | glad - transactions - today - invest - company | 17 | 5698_glad_transactions_today_invest |
| 5699 | rate - country - business - benefits - latest | 17 | 5699_rate_country_business_benefits |
| 5700 | ai - new - - - | 17 | 5700_ai_new__ |
| 5701 | ama - type - dear - digital - user | 17 | 5701_ama_type_dear_digital |
| 5702 | moving - key - action - support - price | 17 | 5702_moving_key_action_support |
| 5703 | word - wallet - note - - | 17 | 5703_word_wallet_note_ |
| 5704 | chat - ama - question - update - rewards | 17 | 5704_chat_ama_question_update |
| 5705 | features - feature - difference - person - option | 17 | 5705_features_feature_difference_person |
| 5706 | rate - low - exchange - account - play | 17 | 5706_rate_low_exchange_account |
| 5707 | holders - smart - rewards - features - chain | 17 | 5707_holders_smart_rewards_features |
| 5708 | rate - bitcoin - action - change - price | 17 | 5708_rate_bitcoin_action_change |
| 5709 | example - yesterday - moment - alot - total | 17 | 5709_example_yesterday_moment_alot |
| 5710 | email - account - change - brother - example | 17 | 5710_email_account_change_brother |
| 5711 | previous - attention - kind - assets - data | 17 | 5711_previous_attention_kind_assets |
| 5712 | lost - fake - huge - opportunity - lot | 17 | 5712_lost_fake_huge_opportunity |
| 5713 | way - - - - | 17 | 5713_way___ |
| 5714 | risk - week - best - platform - link | 17 | 5714_risk_week_best_platform |
| 5715 | increase - pay - term - site - rate | 17 | 5715_increase_pay_term_site |
| 5716 | family - digital - - - | 17 | 5716_family_digital__ |
| 5717 | life - channel - - - | 17 | 5717_life_channel__ |
| 5718 | true - open - meet - dm - real | 17 | 5718_true_open_meet_dm |
| 5719 | night - man - thing - bro - way | 17 | 5719_night_man_thing_bro |
| 5720 | - - - - | 17 | 5720____ |
| 5721 | period - - - - | 17 | 5721_period___ |
| 5722 | - - - - | 17 | 5722____ |
| 5723 | tbh - team - words - idea - issue | 17 | 5723_tbh_team_words_idea |
| 5724 | digital - profits - deposit - daily - wallet | 17 | 5724_digital_profits_deposit_daily |
| 5725 | receive - minimum - deposit - withdrawal - reward | 17 | 5725_receive_minimum_deposit_withdrawal |
| 5726 | end - mate - buddy - question - bad | 17 | 5726_end_mate_buddy_question |
| 5727 | event - previous - blockchain - product - meet | 17 | 5727_event_previous_blockchain_product |
| 5728 | transactions - wallet - company - - | 17 | 5728_transactions_wallet_company_ |
| 5729 | available - code - business - number - sure | 17 | 5729_available_code_business_number |
| 5730 | search - massive - latest - private - launch | 17 | 5730_search_massive_latest_private |
| 5731 | community - attention - games - ready - amazing | 17 | 5731_community_attention_games_ready |
| 5732 | fees - - - - | 17 | 5732_fees___ |
| 5733 | chat - feature - reply - appreciate - past | 17 | 5733_chat_feature_reply_appreciate |
| 5734 | position - volume - market - price - trading | 17 | 5734_position_volume_market_price |
| 5735 | price - good - key - - | 17 | 5735_price_good_key_ |
| 5736 | type - use - sir - share - lot | 17 | 5736_type_use_sir_share |
| 5737 | ecosystem - stuff - real - world - | 17 | 5737_ecosystem_stuff_real_world |
| 5738 | digital - minutes - days - smart - user | 17 | 5738_digital_minutes_days_smart |
| 5739 | excited - tomorrow - happy - year - lot | 17 | 5739_excited_tomorrow_happy_year |
| 5740 | - - - - | 17 | 5740____ |
| 5741 | chain - similar - tokens - version - token | 17 | 5741_chain_similar_tokens_version |
| 5742 | reply - utc - shit - fun - need | 17 | 5742_reply_utc_shit_fun |
| 5743 | buddy - right - glad - excited - true | 17 | 5743_buddy_right_glad_excited |
| 5744 | open - wallet - update - marketing - opportunity | 17 | 5744_open_wallet_update_marketing |
| 5745 | open - deposit - sir - option - certain | 17 | 5745_open_deposit_sir_option |
| 5746 | - - - - | 17 | 5746____ |
| 5747 | night - positive - life - good - | 17 | 5747_night_positive_life_good |
| 5748 | previous - upcoming - attention - massive - assets | 17 | 5748_previous_upcoming_attention_massive |
| 5749 | ecosystem - users - tokens - app - order | 17 | 5749_ecosystem_users_tokens_app |
| 5750 | positive - strong - buying - team - day | 17 | 5750_positive_strong_buying_team |
| 5751 | details - event - tomorrow - transaction - links | 17 | 5751_details_event_tomorrow_transaction |
| 5752 | times - question - good - - | 17 | 5752_times_question_good_ |
| 5753 | chain - launch - couple - week - announcement | 17 | 5753_chain_launch_couple_week |
| 5754 | form - home - correct - rules - hand | 17 | 5754_form_home_correct_rules |
| 5755 | chance - luck - opportunity - good - | 17 | 5755_chance_luck_opportunity_good |
| 5756 | message - kindly - sir - morning - question | 17 | 5756_message_kindly_sir_morning |
| 5757 | active - fast - available - - | 17 | 5757_active_fast_available_ |
| 5758 | site - dear - chance - open - rules | 17 | 5758_site_dear_chance_open |
| 5759 | recent - crypto - tomorrow - today - | 17 | 5759_recent_crypto_tomorrow_today |
| 5760 | happy - great - space - nice - | 17 | 5760_happy_great_space_nice |
| 5761 | coin - hours - new - - | 17 | 5761_coin_hours_new_ |
| 5762 | easy - great - available - time - | 17 | 5762_easy_great_available_time |
| 5763 | different - previous - total - price - higher | 17 | 5763_different_previous_total_price |
| 5764 | public - - - - | 17 | 5764_public___ |
| 5765 | account - price - accounts - private - reason | 17 | 5765_account_price_accounts_private |
| 5766 | volume - feature - chain - multiple - process | 17 | 5766_volume_feature_chain_multiple |
| 5767 | games - couple - game - upcoming - kind | 17 | 5767_games_couple_game_upcoming |
| 5768 | assets - digital - financial - capital - technology | 17 | 5768_assets_digital_financial_capital |
| 5769 | pool - marketing - plan - development - data | 17 | 5769_pool_marketing_plan_development |
| 5770 | trade - - - - | 17 | 5770_trade___ |
| 5771 | product - strong - order - community - project | 17 | 5771_product_strong_order_community |
| 5772 | service - network - difference - bitcoin - single | 17 | 5772_service_network_difference_bitcoin |
| 5773 | game - blockchain - new - - | 17 | 5773_game_blockchain_new_ |
| 5774 | address - email - mate - case - real | 17 | 5774_address_email_mate_case |
| 5775 | unique - dev - discord - questions - network | 17 | 5775_unique_dev_discord_questions |
| 5776 | - - - - | 17 | 5776____ |
| 5777 | protocol - glad - games - ecosystem - build | 17 | 5777_protocol_glad_games_ecosystem |
| 5778 | pm - - - - | 17 | 5778_pm___ |
| 5779 | job - future - - - | 17 | 5779_job_future__ |
| 5780 | security - admin - risk - group - support | 17 | 5780_security_admin_risk_group |
| 5781 | binance - payment - report - position - important | 17 | 5781_binance_payment_report_position |
| 5782 | bro - fine - good - little - | 17 | 5782_bro_fine_good_little |
| 5783 | details - group - link - information - contact | 17 | 5783_details_group_link_information |
| 5784 | glad - day - man - thanks - friends | 17 | 5784_glad_day_man_thanks |
| 5785 | fees - withdrawal - fast - service - fee | 17 | 5785_fees_withdrawal_fast_service |
| 5786 | experience - meet - couple - real - interested | 17 | 5786_experience_meet_couple_real |
| 5787 | game - - - - | 17 | 5787_game___ |
| 5788 | days - - - - | 17 | 5788_days___ |
| 5789 | transaction - funds - payment - active - deposit | 17 | 5789_transaction_funds_payment_active |
| 5790 | friend - person - fake - thing - right | 17 | 5790_friend_person_fake_thing |
| 5791 | assets - research - talk - development - smart | 17 | 5791_assets_research_talk_development |
| 5792 | utc - version - link - volume - data | 17 | 5792_utc_version_link_volume |
| 5793 | positive - comment - alot - excited - attention | 17 | 5793_positive_comment_alot_excited |
| 5794 | solution - bitcoin - - - | 17 | 5794_solution_bitcoin__ |
| 5795 | cryptocurrency - exchange - future - trading - | 17 | 5795_cryptocurrency_exchange_future_trading |
| 5796 | shit - lots - wrong - reason - action | 17 | 5796_shit_lots_wrong_reason |
| 5797 | - - - - | 17 | 5797____ |
| 5798 | question - - - - | 17 | 5798_question___ |
| 5799 | wallet - time - - - | 17 | 5799_wallet_time__ |
| 5800 | unique - chain - supply - total - ecosystem | 17 | 5800_unique_chain_supply_total |
| 5801 | bad - reply - good - multiple - bit | 17 | 5801_bad_reply_good_multiple |
| 5802 | love - word - case - - | 17 | 5802_love_word_case_ |
| 5803 | liquidity - volume - multiple - tokens - pool | 17 | 5803_liquidity_volume_multiple_tokens |
| 5804 | point - accounts - nice - account - available | 17 | 5804_point_accounts_nice_account |
| 5805 | important - different - process - - | 17 | 5805_important_different_process_ |
| 5806 | project - projects - certain - launch - year | 17 | 5806_project_projects_certain_launch |
| 5807 | bad - trading - - - | 17 | 5807_bad_trading__ |
| 5808 | current - - - - | 17 | 5808_current___ |
| 5809 | - - - - | 17 | 5809____ |
| 5810 | old - years - people - - | 17 | 5810_old_years_people_ |
| 5811 | day - link - - - | 17 | 5811_day_link__ |
| 5812 | digital - - - - | 17 | 5812_digital___ |
| 5813 | - - - - | 17 | 5813____ |
| 5814 | development - - - - | 17 | 5814_development___ |
| 5815 | order - certain - word - value - cryptocurrency | 17 | 5815_order_certain_word_value |
| 5816 | utc - dear - details - trading - users | 17 | 5816_utc_dear_details_trading |
| 5817 | games - chart - funds - holders - website | 17 | 5817_games_chart_funds_holders |
| 5818 | second - right - telegram - free - time | 17 | 5818_second_right_telegram_free |
| 5819 | token - number - report - tokens - update | 17 | 5819_token_number_report_tokens |
| 5820 | months - supply - reply - user - mean | 17 | 5820_months_supply_reply_user |
| 5821 | buying - deposit - minimum - fees - available | 17 | 5821_buying_deposit_minimum_fees |
| 5822 | ta - longer - perfect - phone - term | 17 | 5822_ta_longer_perfect_phone |
| 5823 | cause - announcement - users - rest - main | 17 | 5823_cause_announcement_users_rest |
| 5824 | version - app - web - telegram - phone | 17 | 5824_version_app_web_telegram |
| 5825 | ecosystem - - - - | 17 | 5825_ecosystem___ |
| 5826 | ways - correct - guy - cool - main | 17 | 5826_ways_correct_guy_cool |
| 5827 | - - - - | 17 | 5827____ |
| 5828 | - - - - | 17 | 5828____ |
| 5829 | different - exchange - price - example - past | 17 | 5829_different_exchange_price_example |
| 5830 | event - yesterday - hand - visit - public | 17 | 5830_event_yesterday_hand_visit |
| 5831 | interested - legit - friends - profit - company | 17 | 5831_interested_legit_friends_profit |
| 5832 | cryptocurrency - users - best - - | 17 | 5832_cryptocurrency_users_best_ |
| 5833 | link - years - binance - - | 17 | 5833_link_years_binance_ |
| 5834 | multiple - data - liquidity - btc - investors | 17 | 5834_multiple_data_liquidity_btc |
| 5835 | update - dev - app - guide - upcoming | 17 | 5835_update_dev_app_guide |
| 5836 | comment - rest - positive - unique - development | 17 | 5836_comment_rest_positive_unique |
| 5837 | power - - - - | 17 | 5837_power___ |
| 5838 | - - - - | 17 | 5838____ |
| 5839 | specific - services - interested - questions - amazing | 17 | 5839_specific_services_interested_questions |
| 5840 | ai - pepe - tho - shows - correct | 17 | 5840_ai_pepe_tho_shows |
| 5841 | discord - fun - utc - community - tomorrow | 17 | 5841_discord_fun_utc_community |
| 5842 | excited - chain - announcement - date - tokens | 17 | 5842_excited_chain_announcement_date |
| 5843 | home - company - dm - message - | 17 | 5843_home_company_dm_message |
| 5844 | things - week - binance - bitcoin - | 17 | 5844_things_week_binance_bitcoin |
| 5845 | ethereum - major - case - value - real | 17 | 5845_ethereum_major_case_value |
| 5846 | current - best - need - experience - thanks | 17 | 5846_current_best_need_experience |
| 5847 | simple - thought - people - good - today | 17 | 5847_simple_thought_people_good |
| 5848 | phone - - - - | 17 | 5848_phone___ |
| 5849 | matter - bit - current - point - chart | 17 | 5849_matter_bit_current_point |
| 5850 | games - chance - utc - period - pool | 17 | 5850_games_chance_utc_period |
| 5851 | positive - clear - major - agree - option | 17 | 5851_positive_clear_major_agree |
| 5852 | global - new - - - | 17 | 5852_global_new__ |
| 5853 | night - ready - dude - dm - game | 17 | 5853_night_ready_dude_dm |
| 5854 | super - rate - times - service - services | 17 | 5854_super_rate_times_service |
| 5855 | deal - report - swap - possible - | 17 | 5855_deal_report_swap_possible |
| 5856 | key - recent - announcement - experience - guys | 17 | 5856_key_recent_announcement_experience |
| 5857 | old - accounts - sell - deal - account | 17 | 5857_old_accounts_sell_deal |
| 5858 | certain - reason - cool - year - course | 17 | 5858_certain_reason_cool_year |
| 5859 | version - ethereum - - - | 17 | 5859_version_ethereum__ |
| 5860 | past - position - week - better - price | 17 | 5860_past_position_week_better |
| 5861 | early - contract - secure - power - opportunity | 17 | 5861_early_contract_secure_power |
| 5862 | opportunity - thanks - great - biggest - excited | 17 | 5862_opportunity_thanks_great_biggest |
| 5863 | period - details - pm - short - small | 17 | 5863_period_details_pm_short |
| 5864 | - - - - | 17 | 5864____ |
| 5865 | success - hard - work - - | 17 | 5865_success_hard_work_ |
| 5866 | play - btc - price - bitcoin - | 17 | 5866_play_btc_price_bitcoin |
| 5867 | better - - - - | 17 | 5867_better___ |
| 5868 | page - moment - open - step - power | 17 | 5868_page_moment_open_step |
| 5869 | question - - - - | 17 | 5869_question___ |
| 5870 | family - close - lots - amazing - members | 17 | 5870_family_close_lots_amazing |
| 5871 | couple - meet - interested - real - ok | 17 | 5871_couple_meet_interested_real |
| 5872 | hours - new - days - - | 17 | 5872_hours_new_days_ |
| 5873 | plans - future - blockchain - new - | 17 | 5873_plans_future_blockchain_new |
| 5874 | crypto - bitcoin - - - | 17 | 5874_crypto_bitcoin__ |
| 5875 | free - company - investment - - | 17 | 5875_free_company_investment_ |
| 5876 | active - page - big - - | 17 | 5876_active_page_big_ |
| 5877 | account - guy - night - person - lol | 17 | 5877_account_guy_night_person |
| 5878 | period - daily - minimum - liquidity - safe | 17 | 5878_period_daily_minimum_liquidity |
| 5879 | issue - fix - sorry - team - issues | 17 | 5879_issue_fix_sorry_team |
| 5880 | btc - price - - - | 17 | 5880_btc_price__ |
| 5881 | welcome - questions - glad - community - latest | 17 | 5881_welcome_questions_glad_community |
| 5882 | thinking - positive - life - face - morning | 17 | 5882_thinking_positive_life_face |
| 5883 | deal - says - blockchain - assets - issue | 17 | 5883_deal_says_blockchain_assets |
| 5884 | years - free - bitcoin - technology - capital | 17 | 5884_years_free_bitcoin_technology |
| 5885 | trade - link - chain - order - action | 17 | 5885_trade_link_chain_order |
| 5886 | possible - guys - chain - potential - team | 17 | 5886_possible_guys_chain_potential |
| 5887 | half - close - hours - hour - sorry | 17 | 5887_half_close_hours_hour |
| 5888 | ways - game - chain - new - | 17 | 5888_ways_game_chain_new |
| 5889 | business - services - service - - | 17 | 5889_business_services_service_ |
| 5890 | main - rules - morning - thing - today | 17 | 5890_main_rules_morning_thing |
| 5891 | luck - week - live - community - good | 17 | 5891_luck_week_live_community |
| 5892 | stuff - things - - - | 17 | 5892_stuff_things__ |
| 5893 | shit - ones - cool - friend - address | 17 | 5893_shit_ones_cool_friend |
| 5894 | report - - - - | 17 | 5894_report___ |
| 5895 | block - important - fake - safe - support | 17 | 5895_block_important_fake_safe |
| 5896 | morning - crypto - nice - buddy - night | 17 | 5896_morning_crypto_nice_buddy |
| 5897 | family - man - year - - | 17 | 5897_family_man_year_ |
| 5898 | contact - available - - - | 17 | 5898_contact_available__ |
| 5899 | different - platform - legit - follow - real | 17 | 5899_different_platform_legit_follow |
| 5900 | moment - global - - - | 17 | 5900_moment_global__ |
| 5901 | report - personal - rate - funds - information | 17 | 5901_report_personal_rate_funds |
| 5902 | binance - payment - money - interesting - focus | 17 | 5902_binance_payment_money_interesting |
| 5903 | global - future - ready - price - network | 17 | 5903_global_future_ready_price |
| 5904 | day - - - - | 17 | 5904_day___ |
| 5905 | mate - today - sir - man - dm | 17 | 5905_mate_today_sir_man |
| 5906 | luck - event - thanks - message - website | 17 | 5906_luck_event_thanks_message |
| 5907 | members - free - minutes - users - weeks | 17 | 5907_members_free_minutes_users |
| 5908 | things - life - - - | 17 | 5908_things_life__ |
| 5909 | technology - - - - | 17 | 5909_technology___ |
| 5910 | nice - wrong - dear - day - happy | 17 | 5910_nice_wrong_dear_day |
| 5911 | event - transaction - users - new - | 17 | 5911_event_transaction_users_new |
| 5912 | latest - risk - better - - | 17 | 5912_latest_risk_better_ |
| 5913 | person - email - contact - information - address | 17 | 5913_person_email_contact_information |
| 5914 | opinion - major - following - attention - research | 17 | 5914_opinion_major_following_attention |
| 5915 | hand - team - stuff - details - sir | 17 | 5915_hand_team_stuff_details |
| 5916 | win - risk - level - official - daily | 17 | 5916_win_risk_level_official |
| 5917 | free - dm - quick - interested - fast | 17 | 5917_free_dm_quick_interested |
| 5918 | fun - - - - | 17 | 5918_fun___ |
| 5919 | dear - perfect - chain - night - announcement | 17 | 5919_dear_perfect_chain_night |
| 5920 | liquidity - similar - fine - bit - current | 17 | 5920_liquidity_similar_fine_bit |
| 5921 | successful - days - best - just - platform | 17 | 5921_successful_days_best_just |
| 5922 | ai - sense - interesting - word - network | 17 | 5922_ai_sense_interesting_word |
| 5923 | new - - - - | 17 | 5923_new___ |
| 5924 | accounts - discord - reward - account - telegram | 17 | 5924_accounts_discord_reward_account |
| 5925 | moment - ones - option - issues - point | 17 | 5925_moment_ones_option_issues |
| 5926 | worth - today - - - | 17 | 5926_worth_today__ |
| 5927 | year - investors - trading - platform - bitcoin | 17 | 5927_year_investors_trading_platform |
| 5928 | hours - link - company - - | 17 | 5928_hours_link_company_ |
| 5929 | - - - - | 17 | 5929____ |
| 5930 | major - key - security - head - network | 17 | 5930_major_key_security_head |
| 5931 | web - current - smart - safe - chart | 17 | 5931_web_current_smart_safe |
| 5932 | worth - funds - - - | 17 | 5932_worth_funds__ |
| 5933 | level - - - - | 17 | 5933_level___ |
| 5934 | launch - - - - | 17 | 5934_launch___ |
| 5935 | market - good - today - - | 17 | 5935_market_good_today_ |
| 5936 | lot - use - opportunity - - | 17 | 5936_lot_use_opportunity_ |
| 5937 | order - trade - link - - | 17 | 5937_order_trade_link_ |
| 5938 | - - - - | 17 | 5938____ |
| 5939 | excited - similar - video - interesting - worth | 17 | 5939_excited_similar_video_interesting |
| 5940 | - - - - | 17 | 5940____ |
| 5941 | small - release - public - future - crypto | 17 | 5941_small_release_public_future |
| 5942 | free - second - ready - - | 17 | 5942_free_second_ready_ |
| 5943 | volume - total - market - update - win | 17 | 5943_volume_total_market_update |
| 5944 | lot - assets - right - trading - lots | 17 | 5944_lot_assets_right_trading |
| 5945 | dude - dev - guy - case - drop | 17 | 5945_dude_dev_guy_case |
| 5946 | - - - - | 17 | 5946____ |
| 5947 | high - bitcoin - new - - | 17 | 5947_high_bitcoin_new_ |
| 5948 | morning - chat - tomorrow - discord - talk | 17 | 5948_morning_chat_tomorrow_discord |
| 5949 | attention - exciting - chart - chance - website | 17 | 5949_attention_exciting_chart_chance |
| 5950 | - - - - | 17 | 5950____ |
| 5951 | investors - opportunity - money - good - | 17 | 5951_investors_opportunity_money_good |
| 5952 | hope - yesterday - positive - bit - high | 17 | 5952_hope_yesterday_positive_bit |
| 5953 | run - btc - price - bitcoin - lol | 17 | 5953_run_btc_price_bitcoin |
| 5954 | nft - transactions - token - - | 16 | 5954_nft_transactions_token_ |
| 5955 | chart - data - available - issues - personal | 16 | 5955_chart_data_available_issues |
| 5956 | capital - - - - | 16 | 5956_capital___ |
| 5957 | action - hard - btc - price - multiple | 16 | 5957_action_hard_btc_price |
| 5958 | ready - idea - - - | 16 | 5958_ready_idea__ |
| 5959 | btc - price - close - alot - deal | 16 | 5959_btc_price_close_alot |
| 5960 | low - price - - - | 16 | 5960_low_price__ |
| 5961 | end - - - - | 16 | 5961_end___ |
| 5962 | - - - - | 16 | 5962____ |
| 5963 | luck - man - sir - sure - way | 16 | 5963_luck_man_sir_sure |
| 5964 | game - discord - private - easy - order | 16 | 5964_game_discord_private_easy |
| 5965 | past - space - week - experience - | 16 | 5965_past_space_week_experience |
| 5966 | sense - version - times - place - things | 16 | 5966_sense_version_times_place |
| 5967 | win - risk - level - official - daily | 16 | 5967_win_risk_level_official |
| 5968 | - - - - | 16 | 5968____ |
| 5969 | claim - rewards - transactions - tokens - | 16 | 5969_claim_rewards_transactions_tokens |
| 5970 | assets - plans - access - new - platform | 16 | 5970_assets_plans_access_new |
| 5971 | capital - - - - | 16 | 5971_capital___ |
| 5972 | action - change - price - - | 16 | 5972_action_change_price_ |
| 5973 | - - - - | 16 | 5973____ |
| 5974 | plans - services - ai - - | 16 | 5974_plans_services_ai_ |
| 5975 | day - - - - | 16 | 5975_day___ |
| 5976 | lot - market - glad - - | 16 | 5976_lot_market_glad_ |
| 5977 | btc - close - data - term - price | 16 | 5977_btc_close_data_term |
| 5978 | half - words - alot - simple - kind | 16 | 5978_half_words_alot_simple |
| 5979 | night - rest - brother - tomorrow - good | 16 | 5979_night_rest_brother_tomorrow |
| 5980 | reward - token - rewards - chance - worth | 16 | 5980_reward_token_rewards_chance |
| 5981 | - - - - | 16 | 5981____ |
| 5982 | late - normal - reply - fix - shit | 16 | 5982_late_normal_reply_fix |
| 5983 | hours - biggest - new - days - | 16 | 5983_hours_biggest_new_days |
| 5984 | - - - - | 16 | 5984____ |
| 5985 | start - - - - | 16 | 5985_start___ |
| 5986 | - - - - | 16 | 5986____ |
| 5987 | launch - sense - network - weeks - upcoming | 16 | 5987_launch_sense_network_weeks |
| 5988 | volume - access - price - low - short | 16 | 5988_volume_access_price_low |
| 5989 | listing - sign - win - nft - utc | 16 | 5989_listing_sign_win_nft |
| 5990 | words - man - day - people - time | 16 | 5990_words_man_day_people |
| 5991 | ama - opinion - - - | 16 | 5991_ama_opinion__ |
| 5992 | issues - words - days - - | 16 | 5992_issues_words_days_ |
| 5993 | links - rules - code - members - selling | 16 | 5993_links_rules_code_members |
| 5994 | drop - massive - place - post - pepe | 16 | 5994_drop_massive_place_post |
| 5995 | little - life - excited - message - business | 16 | 5995_little_life_excited_message |
| 5996 | higher - pool - share - better - awesome | 16 | 5996_higher_pool_share_better |
| 5997 | wallet - - - - | 16 | 5997_wallet___ |
| 5998 | late - single - today - platform - awesome | 16 | 5998_late_single_today_platform |
| 5999 | dear - participate - second - strong - details | 16 | 5999_dear_participate_second_strong |
| 6000 | break - private - day - - | 16 | 6000_break_private_day_ |
| 6001 | report - coin - bitcoin - - | 16 | 6001_report_coin_bitcoin_ |
| 6002 | happy - - - - | 16 | 6002_happy___ |
| 6003 | short - bitcoin - btc - - | 16 | 6003_short_bitcoin_btc_ |
| 6004 | right - things - trust - main - thing | 16 | 6004_right_things_trust_main |
| 6005 | dude - normal - night - friend - mate | 16 | 6005_dude_normal_night_friend |
| 6006 | product - development - features - user - latest | 16 | 6006_product_development_features_user |
| 6007 | fact - little - fee - wrong - family | 16 | 6007_fact_little_fee_wrong |
| 6008 | sign - future - chat - telegram - group | 16 | 6008_sign_future_chat_telegram |
| 6009 | strong - services - fees - service - life | 16 | 6009_strong_services_fees_service |
| 6010 | holders - chart - true - launch - contract | 16 | 6010_holders_chart_true_launch |
| 6011 | public - ai - use - new - | 16 | 6011_public_ai_use_new |
| 6012 | idea - - - - | 16 | 6012_idea___ |
| 6013 | marketing - easy - massive - fast - dev | 16 | 6013_marketing_easy_massive_fast |
| 6014 | global - months - information - - | 16 | 6014_global_months_information_ |
| 6015 | nice - cool - good - - | 16 | 6015_nice_cool_good_ |
| 6016 | funds - - - - | 16 | 6016_funds___ |
| 6017 | opinion - couple - assets - exchange - channel | 16 | 6017_opinion_couple_assets_exchange |
| 6018 | run - space - simple - attention - idea | 16 | 6018_run_space_simple_attention |
| 6019 | term - fun - space - plans - long | 16 | 6019_term_fun_space_plans |
| 6020 | run - normal - solution - idea - code | 16 | 6020_run_normal_solution_idea |
| 6021 | real - way - sign - - | 16 | 6021_real_way_sign_ |
| 6022 | couple - guy - ones - power - place | 16 | 6022_couple_guy_ones_power |
| 6023 | product - close - open - fun - volume | 16 | 6023_product_close_open_fun |
| 6024 | liquidity - shit - positive - protocol - volume | 16 | 6024_liquidity_shit_positive_protocol |
| 6025 | past - ecosystem - year - team - | 16 | 6025_past_ecosystem_year_team |
| 6026 | fake - users - email - services - cryptocurrency | 16 | 6026_fake_users_email_services |
| 6027 | chance - exciting - development - community - tokens | 16 | 6027_chance_exciting_development_community |
| 6028 | media - news - days - token - report | 16 | 6028_media_news_days_token |
| 6029 | volume - token - trading - tokens - new | 16 | 6029_volume_token_trading_tokens |
| 6030 | technology - security - ethereum - network - new | 16 | 6030_technology_security_ethereum_network |
| 6031 | assets - digital - times - - | 16 | 6031_assets_digital_times_ |
| 6032 | rules - digital - assets - major - details | 16 | 6032_rules_digital_assets_major |
| 6033 | profit - cool - tg - order - thanks | 16 | 6033_profit_cool_tg_order |
| 6034 | opinion - fact - yesterday - answer - wrong | 16 | 6034_opinion_fact_yesterday_answer |
| 6035 | person - guys - people - - | 16 | 6035_person_guys_people_ |
| 6036 | successful - thanks - withdrawal - good - channel | 16 | 6036_successful_thanks_withdrawal_good |
| 6037 | pool - rewards - total - reward - details | 16 | 6037_pool_rewards_total_reward |
| 6038 | solution - lol - man - help - time | 16 | 6038_solution_lol_man_help |
| 6039 | links - search - sorry - words - telegram | 16 | 6039_links_search_sorry_words |
| 6040 | specific - development - set - fair - dev | 16 | 6040_specific_development_set_fair |
| 6041 | pepe - holders - rewards - day - | 16 | 6041_pepe_holders_rewards_day |
| 6042 | hope - better - fine - break - tho | 16 | 6042_hope_better_fine_break |
| 6043 | months - transactions - fees - app - small | 16 | 6043_months_transactions_fees_app |
| 6044 | fun - dude - difference - - | 16 | 6044_fun_dude_difference_ |
| 6045 | control - major - recent - security - global | 16 | 6045_control_major_recent_security |
| 6046 | special - tho - country - course - way | 16 | 6046_special_tho_country_course |
| 6047 | ethereum - chain - smart - similar - space | 16 | 6047_ethereum_chain_smart_similar |
| 6048 | contact - active - page - available - | 16 | 6048_contact_active_page_available |
| 6049 | world - day - - - | 16 | 6049_world_day__ |
| 6050 | problems - period - short - amazing - successful | 16 | 6050_problems_period_short_amazing |
| 6051 | pm - minimum - details - easy - contact | 16 | 6051_pm_minimum_details_easy |
| 6052 | buddy - alright - mate - awesome - shit | 16 | 6052_buddy_alright_mate_awesome |
| 6053 | week - ask - twitter - big - great | 16 | 6053_week_ask_twitter_big |
| 6054 | morning - cool - today - week - great | 16 | 6054_morning_cool_today_week |
| 6055 | problem - launch - - - | 16 | 6055_problem_launch__ |
| 6056 | links - personal - dear - chat - members | 16 | 6056_links_personal_dear_chat |
| 6057 | financial - opportunity - development - cryptocurrency - unique | 16 | 6057_financial_opportunity_development_cryptocurrency |
| 6058 | private - play - ready - launch - token | 16 | 6058_private_play_ready_launch |
| 6059 | plan - profit - digital - days - daily | 16 | 6059_plan_profit_digital_days |
| 6060 | place - space - fast - way - time | 16 | 6060_place_space_fast_way |
| 6061 | point - focus - private - security - night | 16 | 6061_point_focus_private_security |
| 6062 | previous - little - market - big - | 16 | 6062_previous_little_market_big |
| 6063 | rules - transactions - successful - services - success | 16 | 6063_rules_transactions_successful_services |
| 6064 | connect - site - wallet - hand - game | 16 | 6064_connect_site_wallet_hand |
| 6065 | simple - term - reward - claim - long | 16 | 6065_simple_term_reward_claim |
| 6066 | potential - service - similar - normal - turn | 16 | 6066_potential_service_similar_normal |
| 6067 | current - future - price - available - buy | 16 | 6067_current_future_price_available |
| 6068 | rules - assets - data - members - new | 16 | 6068_rules_assets_data_members |
| 6069 | date - months - updates - days - rest | 16 | 6069_date_months_updates_days |
| 6070 | list - correct - latest - app - things | 16 | 6070_list_correct_latest_app |
| 6071 | action - bad - sure - - | 16 | 6071_action_bad_sure_ |
| 6072 | blockchain - - - - | 16 | 6072_blockchain___ |
| 6073 | report - dude - likely - buddy - real | 16 | 6073_report_dude_likely_buddy |
| 6074 | minutes - transactions - trade - - | 16 | 6074_minutes_transactions_trade_ |
| 6075 | week - service - recent - cryptocurrency - news | 16 | 6075_week_service_recent_cryptocurrency |
| 6076 | bitcoin - - - - | 16 | 6076_bitcoin___ |
| 6077 | opinion - issues - different - job - run | 16 | 6077_opinion_issues_different_job |
| 6078 | payment - major - period - digital - data | 16 | 6078_payment_major_period_digital |
| 6079 | year - price - - - | 16 | 6079_year_price__ |
| 6080 | good - - - - | 16 | 6080_good___ |
| 6081 | order - action - group - - | 16 | 6081_order_action_group_ |
| 6082 | control - group - trade - crypto - link | 16 | 6082_control_group_trade_crypto |
| 6083 | major - recent - cryptocurrency - - | 16 | 6083_major_recent_cryptocurrency_ |
| 6084 | supply - fee - contract - total - chart | 16 | 6084_supply_fee_contract_total |
| 6085 | swap - product - fee - chain - pool | 16 | 6085_swap_product_fee_chain |
| 6086 | crazy - marketing - chance - launch - huge | 16 | 6086_crazy_marketing_chance_launch |
| 6087 | trading - - - - | 16 | 6087_trading___ |
| 6088 | extra - thanks - withdrawal - deposit - option | 16 | 6088_extra_thanks_withdrawal_deposit |
| 6089 | - - - - | 16 | 6089____ |
| 6090 | page - research - main - home - site | 16 | 6090_page_research_main_home |
| 6091 | rate - exchange - watch - video - end | 16 | 6091_rate_exchange_watch_video |
| 6092 | company - link - - - | 16 | 6092_company_link__ |
| 6093 | correct - single - key - details - feature | 16 | 6093_correct_single_key_details |
| 6094 | matter - months - - - | 16 | 6094_matter_months__ |
| 6095 | dude - special - shit - attention - type | 16 | 6095_dude_special_shit_attention |
| 6096 | binance - - - - | 16 | 6096_binance___ |
| 6097 | fix - hard - coin - smart - ethereum | 16 | 6097_fix_hard_coin_smart |
| 6098 | access - step - mate - update - rewards | 16 | 6098_access_step_mate_update |
| 6099 | okay - date - open - guy - | 16 | 6099_okay_date_open_guy |
| 6100 | short - mind - total - supply - sell | 16 | 6100_short_mind_total_supply |
| 6101 | massive - ethereum - digital - group - | 16 | 6101_massive_ethereum_digital_group |
| 6102 | country - accounts - telegram - number - old | 16 | 6102_country_accounts_telegram_number |
| 6103 | hand - accounts - possible - available - problems | 16 | 6103_hand_accounts_possible_available |
| 6104 | public - utc - assets - email - worth | 16 | 6104_public_utc_assets_email |
| 6105 | face - fine - kind - assets - share | 16 | 6105_face_fine_kind_assets |
| 6106 | twitter - little - post - work - group | 16 | 6106_twitter_little_post_work |
| 6107 | action - potential - new - step - | 16 | 6107_action_potential_new_step |
| 6108 | product - major - global - capital - research | 16 | 6108_product_major_global_capital |
| 6109 | moment - market - - - | 16 | 6109_moment_market__ |
| 6110 | build - dev - fix - update - user | 16 | 6110_build_dev_fix_update |
| 6111 | night - fun - game - - | 16 | 6111_night_fun_game_ |
| 6112 | special - pool - participate - feature - new | 16 | 6112_special_pool_participate_feature |
| 6113 | code - total - live - - | 16 | 6113_code_total_live_ |
| 6114 | person - attention - phone - future - good | 16 | 6114_person_attention_phone_future |
| 6115 | question - week - live - members - new | 16 | 6115_question_week_live_members |
| 6116 | dm - ama - - - | 16 | 6116_dm_ama__ |
| 6117 | fact - news - moment - huge - cool | 16 | 6117_fact_news_moment_huge |
| 6118 | contract - website - code - - | 16 | 6118_contract_website_code_ |
| 6119 | network - weeks - join - private - future | 16 | 6119_network_weeks_join_private |
| 6120 | data - specific - feature - space - details | 16 | 6120_data_specific_feature_space |
| 6121 | transaction - note - dear - thanks - funds | 16 | 6121_transaction_note_dear_thanks |
| 6122 | protocol - ecosystem - defi - liquidity - assets | 16 | 6122_protocol_ecosystem_defi_liquidity |
| 6123 | answer - problem - admin - private - sir | 16 | 6123_answer_problem_admin_private |
| 6124 | listing - special - main - token - contract | 16 | 6124_listing_special_main_token |
| 6125 | person - luck - good - - | 16 | 6125_person_luck_good_ |
| 6126 | use - - - - | 16 | 6126_use___ |
| 6127 | increase - money - potential - long - great | 16 | 6127_increase_money_potential_long |
| 6128 | global - app - users - legit - links | 16 | 6128_global_app_users_legit |
| 6129 | sure - team - profit - trading - money | 16 | 6129_sure_team_profit_trading |
| 6130 | phone - smart - list - plan - easy | 16 | 6130_phone_smart_list_plan |
| 6131 | life - works - low - level - world | 16 | 6131_life_works_low_level |
| 6132 | rewards - reward - specific - opinion - super | 16 | 6132_rewards_reward_specific_opinion |
| 6133 | technology - ai - questions - - | 16 | 6133_technology_ai_questions_ |
| 6134 | normal - cool - matter - place - hard | 16 | 6134_normal_cool_matter_place |
| 6135 | issues - digital - - - | 16 | 6135_issues_digital__ |
| 6136 | cool - year - good - - | 16 | 6136_cool_year_good_ |
| 6137 | reward - rewards - interesting - shit - reason | 16 | 6137_reward_rewards_interesting_shit |
| 6138 | sense - crazy - moment - country - issues | 16 | 6138_sense_crazy_moment_country |
| 6139 | face - country - easy - search - code | 16 | 6139_face_country_easy_search |
| 6140 | exchange - wallet - link - trading - | 16 | 6140_exchange_wallet_link_trading |
| 6141 | possible - liquidity - bitcoin - - | 16 | 6141_possible_liquidity_bitcoin_ |
| 6142 | app - hand - features - specific - feature | 16 | 6142_app_hand_features_specific |
| 6143 | value - projects - nft - friends - hit | 16 | 6143_value_projects_nft_friends |
| 6144 | search - stuff - cool - type - low | 16 | 6144_search_stuff_cool_type |
| 6145 | user - recent - guy - latest - mate | 16 | 6145_user_recent_guy_latest |
| 6146 | thanks - good - fine - nice - | 16 | 6146_thanks_good_fine_nice |
| 6147 | hand - reply - yesterday - night - kind | 16 | 6147_hand_reply_yesterday_night |
| 6148 | time - - - - | 16 | 6148_time___ |
| 6149 | plan - love - offer - hope - ask | 16 | 6149_plan_love_offer_hope |
| 6150 | buddy - times - lot - thanks - platform | 16 | 6150_buddy_times_lot_thanks |
| 6151 | clear - app - data - change - safe | 16 | 6151_clear_app_data_change |
| 6152 | problems - product - chance - recent - public | 16 | 6152_problems_product_chance_recent |
| 6153 | reply - answer - wrong - friend - help | 16 | 6153_reply_answer_wrong_friend |
| 6154 | - - - - | 16 | 6154____ |
| 6155 | stay - profits - - - | 16 | 6155_stay_profits__ |
| 6156 | friend - country - place - nice - interested | 16 | 6156_friend_country_place_nice |
| 6157 | power - key - life - easy - love | 16 | 6157_power_key_life_easy |
| 6158 | token - strong - guys - potential - project | 16 | 6158_token_strong_guys_potential |
| 6159 | work - community - team - - | 16 | 6159_work_community_team_ |
| 6160 | open - rules - code - app - dear | 16 | 6160_open_rules_code_app |
| 6161 | number - members - telegram - social - free | 16 | 6161_number_members_telegram_social |
| 6162 | thought - crazy - hand - attention - positive | 16 | 6162_thought_crazy_hand_attention |
| 6163 | account - minimum - cryptocurrency - profits - situation | 16 | 6163_account_minimum_cryptocurrency_profits |
| 6164 | interested - - - - | 16 | 6164_interested___ |
| 6165 | deal - old - legit - btc - days | 16 | 6165_deal_old_legit_btc |
| 6166 | buddy - glad - happy - good - amazing | 16 | 6166_buddy_glad_happy_good |
| 6167 | price - bitcoin - action - liquidity - day | 16 | 6167_price_bitcoin_action_liquidity |
| 6168 | links - binance - bitcoin - media - financial | 16 | 6168_links_binance_bitcoin_media |
| 6169 | lots - ways - simple - thing - ready | 16 | 6169_lots_ways_simple_thing |
| 6170 | worth - bitcoin - money - - | 16 | 6170_worth_bitcoin_money_ |
| 6171 | unique - game - digital - years - experience | 16 | 6171_unique_game_digital_years |
| 6172 | project - great - nice - potential - right | 16 | 6172_project_great_nice_potential |
| 6173 | interested - profits - trading - deposit - days | 16 | 6173_interested_profits_trading_deposit |
| 6174 | services - chat - update - contact - offer | 16 | 6174_services_chat_update_contact |
| 6175 | update - marketing - opportunity - wallet - private | 16 | 6175_update_marketing_opportunity_wallet |
| 6176 | guy - new - trust - profits - wallet | 16 | 6176_guy_new_trust_profits |
| 6177 | - - - - | 16 | 6177____ |
| 6178 | private - assets - digital - security - members | 16 | 6178_private_assets_digital_security |
| 6179 | long - trading - country - year - nice | 16 | 6179_long_trading_country_year |
| 6180 | data - ai - protocol - multiple - projects | 16 | 6180_data_ai_protocol_multiple |
| 6181 | digital - binance - group - risk - safe | 16 | 6181_digital_binance_group_risk |
| 6182 | transaction - fee - rewards - users - transactions | 16 | 6182_transaction_fee_rewards_users |
| 6183 | previous - attention - kind - assets - data | 16 | 6183_previous_attention_kind_assets |
| 6184 | enjoy - financial - platform - right - withdrawal | 16 | 6184_enjoy_financial_platform_right |
| 6185 | win - risk - level - official - daily | 16 | 6185_win_risk_level_official |
| 6186 | morning - moment - today - good - time | 16 | 6186_morning_moment_today_good |
| 6187 | product - technology - - - | 16 | 6187_product_technology__ |
| 6188 | moment - sir - - - | 16 | 6188_moment_sir__ |
| 6189 | time - - - - | 16 | 6189_time___ |
| 6190 | hour - available - - - | 16 | 6190_hour_available__ |
| 6191 | bitcoin - - - - | 16 | 6191_bitcoin___ |
| 6192 | trading - - - - | 16 | 6192_trading___ |
| 6193 | home - ways - digital - open - liquidity | 16 | 6193_home_ways_digital_open |
| 6194 | word - words - message - - | 16 | 6194_word_words_message_ |
| 6195 | transaction - cryptocurrency - bitcoin - fee - fees | 16 | 6195_transaction_cryptocurrency_bitcoin_fee |
| 6196 | thing - bad - good - things - sure | 16 | 6196_thing_bad_good_things |
| 6197 | - - - - | 16 | 6197____ |
| 6198 | services - control - family - period - free | 16 | 6198_services_control_family_period |
| 6199 | bitcoin - wallet - - - | 16 | 6199_bitcoin_wallet__ |
| 6200 | experience - brother - hours - interested - contact | 16 | 6200_experience_brother_hours_interested |
| 6201 | game - games - welcome - fair - stuff | 16 | 6201_game_games_welcome_fair |
| 6202 | - - - - | 16 | 6202____ |
| 6203 | interested - short - months - years - channel | 16 | 6203_interested_short_months_years |
| 6204 | pool - reward - block - rest - fees | 16 | 6204_pool_reward_block_rest |
| 6205 | super - app - network - games - user | 16 | 6205_super_app_network_games |
| 6206 | assets - financial - report - - | 16 | 6206_assets_financial_report_ |
| 6207 | rules - binance - new - - | 16 | 6207_rules_binance_new_ |
| 6208 | - - - - | 16 | 6208____ |
| 6209 | home - marketing - work - network - need | 16 | 6209_home_marketing_work_network |
| 6210 | - - - - | 16 | 6210____ |
| 6211 | platform - - - - | 16 | 6211_platform___ |
| 6212 | order - - - - | 16 | 6212_order___ |
| 6213 | short - bitcoin - question - month - btc | 16 | 6213_short_bitcoin_question_month |
| 6214 | network - bitcoin - - - | 16 | 6214_network_bitcoin__ |
| 6215 | token - alot - morning - run - today | 16 | 6215_token_alot_morning_run |
| 6216 | playing - bad - normal - fair - shit | 16 | 6216_playing_bad_normal_fair |
| 6217 | sir - rewards - reward - special - wrong | 16 | 6217_sir_rewards_reward_special |
| 6218 | moment - year - tomorrow - point - old | 15 | 6218_moment_year_tomorrow_point |
| 6219 | sir - coin - listing - links - token | 15 | 6219_sir_coin_listing_links |
| 6220 | group - message - normal - rules - sorry | 15 | 6220_group_message_normal_rules |
| 6221 | binance - attention - telegram - assets - important | 15 | 6221_binance_attention_telegram_assets |
| 6222 | assets - rules - crypto - country - bit | 15 | 6222_assets_rules_crypto_country |
| 6223 | crazy - fun - smart - friend - contract | 15 | 6223_crazy_fun_smart_friend |
| 6224 | total - friends - share - link - user | 15 | 6224_total_friends_share_link |
| 6225 | account - main - friend - message - new | 15 | 6225_account_main_friend_message |
| 6226 | normal - technology - benefits - order - phone | 15 | 6226_normal_technology_benefits_order |
| 6227 | key - control - holders - details - token | 15 | 6227_key_control_holders_details |
| 6228 | chat - normal - multiple - rules - idea | 15 | 6228_chat_normal_multiple_rules |
| 6229 | update - ethereum - important - market - information | 15 | 6229_update_ethereum_important_market |
| 6230 | coins - normal - coin - spot - holders | 15 | 6230_coins_normal_coin_spot |
| 6231 | guy - dude - comment - stuff - higher | 15 | 6231_guy_dude_comment_stuff |
| 6232 | open - fine - morning - problem - times | 15 | 6232_open_fine_morning_problem |
| 6233 | dm - sure - link - - | 15 | 6233_dm_sure_link_ |
| 6234 | technology - blockchain - latest - number - moment | 15 | 6234_technology_blockchain_latest_number |
| 6235 | morning - good - wrong - people - thing | 15 | 6235_morning_good_wrong_people |
| 6236 | attention - interesting - ones - features - site | 15 | 6236_attention_interesting_ones_features |
| 6237 | things - experience - - - | 15 | 6237_things_experience__ |
| 6238 | comment - happy - tomorrow - great - space | 15 | 6238_comment_happy_tomorrow_great |
| 6239 | tbh - close - dev - interested - term | 15 | 6239_tbh_close_dev_interested |
| 6240 | dev - wallet - situation - major - guy | 15 | 6240_dev_wallet_situation_major |
| 6241 | feature - order - users - assets - multiple | 15 | 6241_feature_order_users_assets |
| 6242 | cool - nice - worth - man - | 15 | 6242_cool_nice_worth_man |
| 6243 | brother - close - minutes - person - problem | 15 | 6243_brother_close_minutes_person |
| 6244 | lot - bad - time - - | 15 | 6244_lot_bad_time_ |
| 6245 | success - successful - term - problem - level | 15 | 6245_success_successful_term_problem |
| 6246 | welcome - mate - excited - morning - questions | 15 | 6246_welcome_mate_excited_morning |
| 6247 | alot - building - cause - crazy - course | 15 | 6247_alot_building_cause_crazy |
| 6248 | announcement - updates - question - twitter - channel | 15 | 6248_announcement_updates_question_twitter |
| 6249 | single - crazy - correct - positive - sorry | 15 | 6249_single_crazy_correct_positive |
| 6250 | discord - early - address - wallet - order | 15 | 6250_discord_early_address_wallet |
| 6251 | daily - current - questions - official - long | 15 | 6251_daily_current_questions_official |
| 6252 | news - token - new - platform - | 15 | 6252_news_token_new_platform |
| 6253 | worth - ones - little - lot - hard | 15 | 6253_worth_ones_little_lot |
| 6254 | little - help - lots - investment - successful | 15 | 6254_little_help_lots_investment |
| 6255 | night - care - hand - rest - morning | 15 | 6255_night_care_hand_rest |
| 6256 | unique - visit - exciting - step - rewards | 15 | 6256_unique_visit_exciting_step |
| 6257 | mind - dude - things - benefits - person | 15 | 6257_mind_dude_things_benefits |
| 6258 | sell - listing - low - week - people | 15 | 6258_sell_listing_low_week |
| 6259 | reason - means - yesterday - sir - ones | 15 | 6259_reason_means_yesterday_sir |
| 6260 | announcement - event - media - social - fun | 15 | 6260_announcement_event_media_social |
| 6261 | usdt - says - ethereum - network - issue | 15 | 6261_usdt_says_ethereum_network |
| 6262 | job - market - position - good - time | 15 | 6262_job_market_position_good |
| 6263 | means - tho - months - correct - word | 15 | 6263_means_tho_months_correct |
| 6264 | cool - kind - friend - family - lol | 15 | 6264_cool_kind_friend_family |
| 6265 | answer - reason - question - times - right | 15 | 6265_answer_reason_question_times |
| 6266 | guide - video - friend - links - admin | 15 | 6266_guide_video_friend_links |
| 6267 | order - situation - possible - example - trade | 15 | 6267_order_situation_possible_example |
| 6268 | better - time - world - way - work | 15 | 6268_better_time_world_way |
| 6269 | welcome - space - supply - big - | 15 | 6269_welcome_space_supply_big |
| 6270 | year - reason - problem - team - thing | 15 | 6270_year_reason_problem_team |
| 6271 | normal - working - issue - problem - bad | 15 | 6271_normal_working_issue_problem |
| 6272 | morning - community - good - - | 15 | 6272_morning_community_good_ |
| 6273 | situation - previous - bad - place - good | 15 | 6273_situation_previous_bad_place |
| 6274 | dude - bit - years - friend - dear | 15 | 6274_dude_bit_years_friend |
| 6275 | bad - dude - bro - good - | 15 | 6275_bad_dude_bro_good |
| 6276 | open - recent - stuff - word - simple | 15 | 6276_open_recent_stuff_word |
| 6277 | buddy - today - friend - days - day | 15 | 6277_buddy_today_friend_days |
| 6278 | way - - - - | 15 | 6278_way___ |
| 6279 | man - team - - - | 15 | 6279_man_team__ |
| 6280 | listing - project - list - simple - fine | 15 | 6280_listing_project_list_simple |
| 6281 | code - phone - number - withdraw - wrong | 15 | 6281_code_phone_number_withdraw |
| 6282 | fun - details - dm - - | 15 | 6282_fun_details_dm_ |
| 6283 | different - okay - wrong - people - lol | 15 | 6283_different_okay_wrong_people |
| 6284 | thought - bitcoin - members - service - people | 15 | 6284_thought_bitcoin_members_service |
</details>
## Training hyperparameters
* calculate_probabilities: False
* language: english
* low_memory: False
* min_topic_size: 10
* n_gram_range: (1, 1)
* nr_topics: None
* seed_topic_list: None
* top_n_words: 10
* verbose: True
## Framework versions
* Numpy: 1.22.4
* HDBSCAN: 0.8.29
* UMAP: 0.5.3
* Pandas: 1.5.3
* Scikit-Learn: 1.2.2
* Sentence-transformers: 2.2.2
* Transformers: 4.30.1
* Numba: 0.56.4
* Plotly: 5.13.1
* Python: 3.10.12
|
Ralpher1/Test
|
Ralpher1
| 2023-06-12T20:53:01Z | 0 | 0 | null |
[
"dataset:fka/awesome-chatgpt-prompts",
"region:us"
] | null | 2023-06-12T20:52:38Z |
---
datasets:
- fka/awesome-chatgpt-prompts
---
|
CodyKilpatrick/ppo-Pyramids
|
CodyKilpatrick
| 2023-06-12T20:48:24Z | 2 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"Pyramids",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-Pyramids",
"region:us"
] |
reinforcement-learning
| 2023-06-12T20:47:51Z |
---
library_name: ml-agents
tags:
- Pyramids
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-Pyramids
---
# **ppo** Agent playing **Pyramids**
This is a trained model of a **ppo** agent playing **Pyramids**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: CodyKilpatrick/ppo-Pyramids
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
Sephiroth777/mrchnjab
|
Sephiroth777
| 2023-06-12T20:01:16Z | 29 | 0 |
diffusers
|
[
"diffusers",
"safetensors",
"text-to-image",
"stable-diffusion",
"license:creativeml-openrail-m",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] |
text-to-image
| 2023-06-10T02:36:54Z |
---
license: creativeml-openrail-m
tags:
- text-to-image
- stable-diffusion
---
### mrchnjab Dreambooth model trained by Sephiroth777 with [TheLastBen's fast-DreamBooth](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb) notebook
Test the concept via A1111 Colab [fast-Colab-A1111](https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb)
Sample pictures of this concept:
|
jasio1304/swin-tiny-patch4-window7-224-finetuned-eurosat
|
jasio1304
| 2023-06-12T19:58:00Z | 219 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"swin",
"image-classification",
"generated_from_trainer",
"dataset:imagefolder",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
image-classification
| 2023-06-11T18:51:24Z |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: swin-tiny-patch4-window7-224-finetuned-eurosat
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.9796296296296296
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# swin-tiny-patch4-window7-224-finetuned-eurosat
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0610
- Accuracy: 0.9796
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| 0.2676 | 1.0 | 190 | 0.1383 | 0.9541 |
| 0.1725 | 2.0 | 380 | 0.0839 | 0.9719 |
| 0.1061 | 3.0 | 570 | 0.0610 | 0.9796 |
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
|
ak0601/Story_gen
|
ak0601
| 2023-06-12T19:52:55Z | 0 | 1 | null |
[
"movie",
"synopsis",
"tagging",
"deep learning",
"clasification",
"summarization",
"en",
"license:apache-2.0",
"region:us"
] |
summarization
| 2023-06-12T19:26:44Z |
---
license: apache-2.0
language:
- en
metrics:
- accuracy
pipeline_tag: summarization
tags:
- movie
- synopsis
- tagging
- deep learning
- clasification
---
|
neuralnavigators/ppo-LunarLander-v2
|
neuralnavigators
| 2023-06-12T19:49:25Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T19:49:08Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 272.31 +/- 16.48
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
Olegiy/ppo-LunarLander-v2-rl-course
|
Olegiy
| 2023-06-12T19:34:11Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T16:40:46Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 274.07 +/- 23.83
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
renyulin/q-Taxi-v3
|
renyulin
| 2023-06-12T19:15:55Z | 0 | 0 | null |
[
"Taxi-v3",
"q-learning",
"reinforcement-learning",
"custom-implementation",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T19:15:52Z |
---
tags:
- Taxi-v3
- q-learning
- reinforcement-learning
- custom-implementation
model-index:
- name: q-Taxi-v3
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Taxi-v3
type: Taxi-v3
metrics:
- type: mean_reward
value: 7.56 +/- 2.71
name: mean_reward
verified: false
---
# **Q-Learning** Agent playing1 **Taxi-v3**
This is a trained model of a **Q-Learning** agent playing **Taxi-v3** .
## Usage
```python
model = load_from_hub(repo_id="renyulin/q-Taxi-v3", filename="q-learning.pkl")
# Don't forget to check if you need to add additional attributes (is_slippery=False etc)
env = gym.make(model["env_id"])
```
|
hannatoenbreker/whisper-dutch-small-v2
|
hannatoenbreker
| 2023-06-12T19:13:02Z | 90 | 1 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"whisper",
"automatic-speech-recognition",
"hf-asr-leaderboard",
"generated_from_trainer",
"nl",
"dataset:mozilla-foundation/common_voice_11_0",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2023-06-12T15:02:22Z |
---
language:
- nl
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
model-index:
- name: Whisper Dutch - RTL
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 11.0
type: mozilla-foundation/common_voice_11_0
config: nl
split: test
args: 'config: nl, split: test'
metrics:
- name: Wer
type: wer
value: 87.5
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Whisper Dutch - RTL
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 1.4419
- Wer: 87.5
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:----:|
| 0.0 | 1000.0 | 1000 | 1.4582 | 62.5 |
| 0.0 | 2000.0 | 2000 | 1.4487 | 62.5 |
| 0.0 | 3000.0 | 3000 | 1.4444 | 75.0 |
| 0.0 | 4000.0 | 4000 | 1.4419 | 87.5 |
### Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3
|
Yuyu12347/Testkurumi
|
Yuyu12347
| 2023-06-12T18:50:53Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T18:45:28Z |
---
license: creativeml-openrail-m
---
|
SHENMU007/neunit_BASE_V8.2
|
SHENMU007
| 2023-06-12T18:50:13Z | 75 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"speecht5",
"text-to-audio",
"1.1.0",
"generated_from_trainer",
"zh",
"dataset:facebook/voxpopuli",
"license:mit",
"endpoints_compatible",
"region:us"
] |
text-to-audio
| 2023-06-12T15:58:16Z |
---
language:
- zh
license: mit
tags:
- 1.1.0
- generated_from_trainer
datasets:
- facebook/voxpopuli
model-index:
- name: SpeechT5 TTS Dutch neunit
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# SpeechT5 TTS Dutch neunit
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the VoxPopuli dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
### Training results
### Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3
|
PhysHunter/ippo-LunarLander-v2
|
PhysHunter
| 2023-06-12T18:34:04Z | 0 | 0 | null |
[
"tensorboard",
"LunarLander-v2",
"ppo",
"deep-reinforcement-learning",
"reinforcement-learning",
"custom-implementation",
"deep-rl-course",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T00:20:05Z |
---
tags:
- LunarLander-v2
- ppo
- deep-reinforcement-learning
- reinforcement-learning
- custom-implementation
- deep-rl-course
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 108.04 +/- 19.21
name: mean_reward
verified: false
---
# PPO Agent Playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2.
# Hyperparameters
```python
{'exp_name': 'ppo'
'repo_id': 'PhysHunter/ippo-LunarLander-v2'
'env_id': 'LunarLander-v2'
'learning_rate': 0.001
'seed': 1
'total_timesteps': 1000000
'torch_deterministic': True
'cuda': True
'record_video': None
'num_envs': 16
'num_steps': 1024
'anneal_lr': False
'gae_lambda': 0.98
'gamma': 0.999
'minibatch_size': 4096
'num_epochs': 16
'adv_norm': True
'clip_coef': 0.2
'clip_value_loss': True
'entropy_coef': 0.01
'value_coef': 0.5
'max_grad_norm': 0.5
'target_kl': None
'num_hiddens': 64
'batch_size': 16384}
```
|
setareh1/ppo-LunarLander-v2
|
setareh1
| 2023-06-12T18:26:04Z | 5 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T18:25:45Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 265.19 +/- 12.28
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
yam8572/qrdqn-SpaceInvaders-v5
|
yam8572
| 2023-06-12T18:13:43Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"ALE/SpaceInvaders-v5",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T18:13:17Z |
---
library_name: stable-baselines3
tags:
- ALE/SpaceInvaders-v5
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: QRDQN
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: ALE/SpaceInvaders-v5
type: ALE/SpaceInvaders-v5
metrics:
- type: mean_reward
value: 445.00 +/- 143.04
name: mean_reward
verified: false
---
# **QRDQN** Agent playing **ALE/SpaceInvaders-v5**
This is a trained model of a **QRDQN** agent playing **ALE/SpaceInvaders-v5**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3)
and the [RL Zoo](https://github.com/DLR-RM/rl-baselines3-zoo).
The RL Zoo is a training framework for Stable Baselines3
reinforcement learning agents,
with hyperparameter optimization and pre-trained agents included.
## Usage (with SB3 RL Zoo)
RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo<br/>
SB3: https://github.com/DLR-RM/stable-baselines3<br/>
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib
Install the RL Zoo (with SB3 and SB3-Contrib):
```bash
pip install rl_zoo3
```
```
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo qrdqn --env ALE/SpaceInvaders-v5 -orga yam8572 -f logs/
python -m rl_zoo3.enjoy --algo qrdqn --env ALE/SpaceInvaders-v5 -f logs/
```
If you installed the RL Zoo3 via pip (`pip install rl_zoo3`), from anywhere you can do:
```
python -m rl_zoo3.load_from_hub --algo qrdqn --env ALE/SpaceInvaders-v5 -orga yam8572 -f logs/
python -m rl_zoo3.enjoy --algo qrdqn --env ALE/SpaceInvaders-v5 -f logs/
```
## Training (with the RL Zoo)
```
python -m rl_zoo3.train --algo qrdqn --env ALE/SpaceInvaders-v5 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo qrdqn --env ALE/SpaceInvaders-v5 -f logs/ -orga yam8572
```
## Hyperparameters
```python
OrderedDict([('batch_size', 32),
('buffer_size', 100000),
('env_wrapper',
['stable_baselines3.common.atari_wrappers.AtariWrapper']),
('exploration_final_eps', 0.01),
('exploration_fraction', 0.1),
('frame_stack', 4),
('gradient_steps', 2),
('learning_rate', 0.0001),
('learning_starts', 100000),
('n_envs', 8),
('n_timesteps', 10000000.0),
('optimize_memory_usage', False),
('policy', 'CnnPolicy'),
('target_update_interval', 1000),
('train_freq', 8),
('normalize', False)])
```
# Environment Arguments
```python
{'render_mode': 'rgb_array'}
```
|
hannatoenbreker/whisper-dutch
|
hannatoenbreker
| 2023-06-12T18:10:04Z | 103 | 4 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"whisper",
"automatic-speech-recognition",
"hf-asr-leaderboard",
"generated_from_trainer",
"nl",
"dataset:mozilla-foundation/common_voice_11_0",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2023-05-11T10:19:43Z |
---
language:
- nl
license: apache-2.0
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
model-index:
- name: Whisper Dutch - RTL
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 11.0
type: mozilla-foundation/common_voice_11_0
config: dutch
split: test
args: dutch
metrics:
- name: Wer
type: wer
value: 24.924531584171586
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# Whisper Dutch - RTL
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1895
- Wer: 24.9245
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 1000
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
| 0.1209 | 0.78 | 1000 | 0.1895 | 24.9245 |
### Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3
|
Joelrrdavis/distilbert-base-uncased-finetuned-emotion
|
Joelrrdavis
| 2023-06-12T18:03:59Z | 105 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"distilbert",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2023-06-12T17:21:44Z |
---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: distilbert-base-uncased-finetuned-emotion
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2120
- Accuracy: 0.9295
- F1: 0.9296
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.8213 | 1.0 | 250 | 0.3055 | 0.905 | 0.9025 |
| 0.2434 | 2.0 | 500 | 0.2120 | 0.9295 | 0.9296 |
### Framework versions
- Transformers 4.16.2
- Pytorch 2.0.1+cu118
- Datasets 1.16.1
- Tokenizers 0.13.3
|
s3ah0rse71325/mt5-small-finetuned-amazon-en-es
|
s3ah0rse71325
| 2023-06-12T17:36:25Z | 3 | 0 |
transformers
|
[
"transformers",
"tf",
"mt5",
"text2text-generation",
"generated_from_keras_callback",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2023-06-12T16:16:48Z |
---
license: apache-2.0
tags:
- generated_from_keras_callback
model-index:
- name: s3ah0rse71325/mt5-small-finetuned-amazon-en-es
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# s3ah0rse71325/mt5-small-finetuned-amazon-en-es
This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset.
It achieves the following results on the evaluation set:
- Train Loss: 4.0230
- Validation Loss: 3.3276
- Epoch: 7
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5.6e-05, 'decay_steps': 9672, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: mixed_float16
### Training results
| Train Loss | Validation Loss | Epoch |
|:----------:|:---------------:|:-----:|
| 9.6526 | 4.1209 | 0 |
| 5.7678 | 3.6775 | 1 |
| 5.0403 | 3.5411 | 2 |
| 4.6453 | 3.4436 | 3 |
| 4.3859 | 3.3928 | 4 |
| 4.2074 | 3.3595 | 5 |
| 4.0890 | 3.3319 | 6 |
| 4.0230 | 3.3276 | 7 |
### Framework versions
- Transformers 4.30.1
- TensorFlow 2.12.0
- Datasets 2.12.0
- Tokenizers 0.13.3
|
Benned/Japanese_POP_ArtStyle
|
Benned
| 2023-06-12T17:33:02Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T17:31:02Z |
---
license: creativeml-openrail-m
---
|
grantprice/bert-base-uncased-squad-v1-finetuned-DND
|
grantprice
| 2023-06-12T17:18:56Z | 56 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"bert",
"text-generation",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-10T00:01:14Z |
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: bert-base-uncased-squad-v1-finetuned-DND
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert-base-uncased-squad-v1-finetuned-DND
This model is a fine-tuned version of [csarron/bert-base-uncased-squad-v1](https://huggingface.co/csarron/bert-base-uncased-squad-v1) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3340
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 2.4852 | 1.0 | 572 | 0.8769 |
| 0.5831 | 2.0 | 1144 | 0.4166 |
| 0.3135 | 3.0 | 1716 | 0.3340 |
### Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
|
huggingtweets/sainte_caramel
|
huggingtweets
| 2023-06-12T17:13:07Z | 129 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"huggingtweets",
"en",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T17:12:57Z |
---
language: en
thumbnail: https://github.com/borisdayma/huggingtweets/blob/master/img/logo.png?raw=true
tags:
- huggingtweets
widget:
- text: "My dream is"
---
<div class="inline-flex flex-col" style="line-height: 1.5;">
<div class="flex">
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1666742073888833539/rZoz5ZIm_400x400.jpg')">
</div>
<div
style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')">
</div>
<div
style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')">
</div>
</div>
<div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 AI BOT 🤖</div>
<div style="text-align: center; font-size: 16px; font-weight: 800">𝐂𝐚𝐫𝐚𝐦𝐞𝐥 🧡 2DFD</div>
<div style="text-align: center; font-size: 14px;">@sainte_caramel</div>
</div>
I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets).
Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)!
## How does it work?
The model uses the following pipeline.
To understand how the model was developed, check the [W&B report](https://wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-Model-to-Generate-Tweets--VmlldzoxMTY5MjI).
## Training data
The model was trained on tweets from 𝐂𝐚𝐫𝐚𝐦𝐞𝐥 🧡 2DFD.
| Data | 𝐂𝐚𝐫𝐚𝐦𝐞𝐥 🧡 2DFD |
| --- | --- |
| Tweets downloaded | 2808 |
| Retweets | 478 |
| Short tweets | 303 |
| Tweets kept | 2027 |
[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/lskhbqwr/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
## Training procedure
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @sainte_caramel's tweets.
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/x5m101wv) for full transparency and reproducibility.
At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/x5m101wv/artifacts) is logged and versioned.
## How to use
You can use this model directly with a pipeline for text generation:
```python
from transformers import pipeline
generator = pipeline('text-generation',
model='huggingtweets/sainte_caramel')
generator("My dream is", num_return_sequences=5)
```
## Limitations and bias
The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
In addition, the data present in the user's tweets further affects the text generated by the model.
## About
*Built by Boris Dayma*
[](https://twitter.com/intent/follow?screen_name=borisdayma)
For more details, visit the project repository.
[](https://github.com/borisdayma/huggingtweets)
|
LoaiThamer2/GPT-4
|
LoaiThamer2
| 2023-06-12T17:12:13Z | 0 | 0 | null |
[
"AI",
"OpenAI",
"ChatGPT",
"GPT-4",
"GPT4",
"AI Language Model",
"Helpful Assistant",
"question-answering",
"en",
"ar",
"zh",
"fr",
"pt",
"es",
"ru",
"uk",
"license:cc-by-nc-sa-4.0",
"region:us"
] |
question-answering
| 2023-06-12T16:48:31Z |
---
license: cc-by-nc-sa-4.0
language:
- en
- ar
- zh
- fr
- pt
- es
- ru
- uk
metrics:
- character
pipeline_tag: question-answering
tags:
- AI
- OpenAI
- ChatGPT
- GPT-4
- GPT4
- AI Language Model
- Helpful Assistant
---
|
grwytwe/jyfririit
|
grwytwe
| 2023-06-12T17:08:15Z | 0 | 0 | null |
[
"region:us"
] | null | 2023-06-12T17:07:41Z |
https://colab.research.google.com/drive/1tIJK2c-KnYiSeNGlUwLQ_h4x-P-UmCw0
https://colab.research.google.com/drive/1aAbLpxSrxwGespAWDijAiQE59x8RdTxx
https://colab.research.google.com/drive/1JJAcBdpwsiGb5FOG2_xjXlWLZQ5pvruz
https://colab.research.google.com/drive/1uuja9xHBAQRmJjr0SqWHnVXSyl0WDfeC
https://colab.research.google.com/drive/1K_q5PkLVqHk44ApqBoUM6woI3DCQCjQF
https://colab.research.google.com/drive/1WK7TgXn7GRZgtQ5AF6GbPWHgG9NUE_yf
https://colab.research.google.com/drive/1vHDF2W0II1mXQmar5wDpBWu6zoIQsq_b
https://colab.research.google.com/drive/1myiGaqGb0B_lqiK9b6EdNXAcf7ATCGE4
https://colab.research.google.com/drive/1OprDkPGL5U6Fze9CnjEDjWI-KVDPtM3M
https://colab.research.google.com/drive/14nw57h78l7CdrZNnasAJxOOL_sGbKUMF
https://colab.research.google.com/drive/1lk0ypBSyO9WYMzqOsPPNPT_cZDkFSHUj
https://colab.research.google.com/drive/18WZR-f2XoiNeWJ_PfefFk6HHTgoQFPTC
https://colab.research.google.com/drive/1Miro8NB1j44Sz0Av9SgozSQ8z7Gihtjj
https://colab.research.google.com/drive/1tEdU571kgU5WpAjhEe4Ec_EOKV-bnKna
https://colab.research.google.com/drive/1iozDBwcmxF_jEJXGR3bmUAW7Xb5BF985
https://colab.research.google.com/drive/1WSudqP_HSZpHOmnZghKs89RnUYGVJyYE
https://colab.research.google.com/drive/12kRtskvQP_ptDoMRG89Gwv02cmf5w2iK
https://colab.research.google.com/drive/1-RILr6OeDCsFfY2s88QGeLyJvbGanSsr
https://colab.research.google.com/drive/1wpA3Nu7K_LEBkyYnEU6xqtRmQP5L8O-e
https://colab.research.google.com/drive/1LpGT0UnVCJF4jBotrKXMFWd8tJBvxIQe
https://colab.research.google.com/drive/15cCtcRnWY5QoTP_SBKDKNaS1aJ4vNq9X
https://colab.research.google.com/drive/1bhHh1KAkw8DJCILFJBJBAZZSKfqTuM2x
https://colab.research.google.com/drive/1_JGIDVyHBMzeEYwOYi35o69p1encKd1n
https://colab.research.google.com/drive/1yGuSIGtooOHyjeNPbwHcHyg3N_cZj5KN
https://colab.research.google.com/drive/1upyQKVoMcNOxoH3G9Qk-OkLNbQbcWjkA
https://colab.research.google.com/drive/1TRH5LJ1ilod7au6c-Pw2pxssf8SMeasm
https://colab.research.google.com/drive/1Rs4AQxId9cyzlQzOx4VYMVZNj6qMcy2E
https://colab.research.google.com/drive/1uqAHKsByZ4-KHtNjbGxp1WouQnDgHkyL
https://colab.research.google.com/drive/1-J5inX7iAr0TCq3nmkms9iBRGQPswdqA
https://colab.research.google.com/drive/1omYzaCYBDPsLL_VDfxqYeESaoj6NT3ID
https://colab.research.google.com/drive/1RdBuAu_FnUqf-DmViTRrb5qoTaRyMW0q
https://colab.research.google.com/drive/1itTwfA7oNt-Bn4Rl_t2JtLqLBGj2Rxi2
https://colab.research.google.com/drive/1IYB1zNm1J_q3_PlGBzcOlqonnE6fDxXr
https://colab.research.google.com/drive/1qrO2Okuih_PmI7qwjvKs7Js5NIMN0X7S
https://colab.research.google.com/drive/1rVxcPfNIeBaMSfJwKZI2idNVbjFFsny3
https://colab.research.google.com/drive/1itb3T1FlR6KONpvPFEJnV0djiFY2s8yh
https://colab.research.google.com/drive/1o-Bfzx1_vNg1qBJG5XAbN4rfP4y4bF4f
https://colab.research.google.com/drive/13UncOrkSTbmN7Z4uBvEoArvLZTgq54xg
https://colab.research.google.com/drive/1ToTJJiuXMVZtMkMxez00BVb_tcNuCnrs
https://colab.research.google.com/drive/1KbzSP64v00M3LAdS2b93IZoIB7yFfkhn
https://colab.research.google.com/drive/1HDCsQcOtWRy-3vf3yuHVCZjsZYGzHj4G
https://colab.research.google.com/drive/1vTRyfl8bXKEEhQsB1hxB2ZQFlBimg1LQ
https://colab.research.google.com/drive/1rJbFOF8Bo2SKa9_J94kcooSb-uQ6H2d7
https://colab.research.google.com/drive/1O6uUleeEwAUJ3ef6n9qByLIcBVMbvk2x
https://colab.research.google.com/drive/1-1N_iNXqydfQlqKjVj7AsX5dSFX5spdg
https://colab.research.google.com/drive/1FsDk9N8Vdym94QK3bN_bvZOzJ0fFO2Fu
https://colab.research.google.com/drive/1gyjXsUQHWbvPv3n7S19xlWNg5uz1d4Hd
https://colab.research.google.com/drive/1P6m-MSw5OFYz16pna3dWNuVNWueGSs5n
https://colab.research.google.com/drive/1h16UaqF1u7byBFsK3gkgVzeleAi88pFE
https://colab.research.google.com/drive/10OtyZG22NNcDKzm6UGVmuuLfYCuw-fs1
https://colab.research.google.com/drive/1vkMrQnUsG1ixSqFHy1ktDBnP5YzMmMK-
https://colab.research.google.com/drive/1KgMvgpYun6PGaJSkmCZukSn1TKmr-qZ9
https://colab.research.google.com/drive/1_cJ0VGTTyBQBjHng-YIbtIRLwW_GGk9j
https://colab.research.google.com/drive/1UXK0hZ1MKcEKa74Iubuf9DVriRheXshZ
https://colab.research.google.com/drive/133cQwiJK1xn60rfCc99O_8CURkeiNSIC
https://colab.research.google.com/drive/1khy69Kgr5f681oE8XqGWQDMankpEWKhr
https://colab.research.google.com/drive/1TrC-QjzYYPnxkT_njGhiYhyELx-Dg5re
https://colab.research.google.com/drive/11EgtxjAlibpqs1mMpgDoq2nv5a8w4oS_
https://colab.research.google.com/drive/1Ff6MFRQyUUS8DPbFCsYxcIueVWjtXc88
https://colab.research.google.com/drive/1AR-Hrrm2dk704cuphYJ6iEHcoI4AiOxF
https://colab.research.google.com/drive/1mKl1PldmOj9E4r6IOTobfEPQT4C0nJ_x
https://colab.research.google.com/drive/1v9Uf9FJzgHFlSwlOHBDxzFrcGLyVmwRF
https://colab.research.google.com/drive/12hjJBRcRG2bkBLiqUNHoCELZmJpPBDVH
https://colab.research.google.com/drive/1_QQDpsBkBmVpBT4lz60euvF9VhCbn2LF
https://colab.research.google.com/drive/1FBWbV-4×6bBe5gfMvO3FZSW09ItfymWp
https://colab.research.google.com/drive/1HQcmLx8uwWNAo4zrus7I9UJvU4MLa9J-
https://colab.research.google.com/drive/1dOZVlNChj_667Gh4AIq-COrrAi2kPMc0
https://colab.research.google.com/drive/14iG5y4XFO-6mEjfMkhFfAOeplNt5pddw
https://colab.research.google.com/drive/1u32qCkZiiyehaFlucQyZ1e8UJ6rQivJs
https://colab.research.google.com/drive/11FVGB6GhVZK9748jWNKFS0wtuu9XEYZ-
https://colab.research.google.com/drive/1qdsuEIuYoSNM5xDL54ewAoJNMenlgPQB
https://colab.research.google.com/drive/1a-XvIiDgu5NfDVcl_nAHkzHDTg6VLAuU
https://colab.research.google.com/drive/1P4zOWa-6ogMCsGHJeJ7GTjoYommTIr56
https://colab.research.google.com/drive/1hWqhNHhr5yVjF1h8WNnyWePaRVnNG6_y
https://colab.research.google.com/drive/1pIyY8PCq4xJDS_rBQIgdjzY3NAXdqEnQ
https://colab.research.google.com/drive/1RY8KZ-u6dil9LqQ0_QGw6rnjCU-dCCZE
https://colab.research.google.com/drive/1OG4ZKKRQShzyzGAq9StEx7IU5IGmGyjA
https://colab.research.google.com/drive/1L24LO8wI9txBdTRDTv13q4qUjo_kcBmH
https://colab.research.google.com/drive/1EPegWzb77f-EIdj4Xf6IiDGj06lZ9qd_
https://colab.research.google.com/drive/1HD6mWgj6BJigTaVHMRVynd3Nx0G4vJpM
https://colab.research.google.com/drive/1IdIIkmVC2KryS_vvkMdNm2kKsbF3blaB
https://colab.research.google.com/drive/1sW9GuF2sQOpKailFBKmo5P9pes5zUejC
https://colab.research.google.com/drive/1VYM7Dx-UPaSKV_7eVHRtPhj94n4xkrCu
https://colab.research.google.com/drive/1BaY0F9E3o3peoKnjwX8OLLEncYEUBCv7
https://colab.research.google.com/drive/1PwjbeDHWkGu8nU2yJkS6iI-icAC14qWH
https://colab.research.google.com/drive/1RgcYCOlDdntcyIQz5oS_fJhEOw4DVwfH
https://colab.research.google.com/drive/1l_RW-w1sMgRFDivycArhNQ-GHVbXDE7j
https://colab.research.google.com/drive/19Js3KCW5RI6wULfhTW7P7J30IVJcelsz
https://colab.research.google.com/drive/1u_F01-535qrrZNbDwdQSEb-1jkIoF3qA
https://colab.research.google.com/drive/1v19J9u7KR6xY0sbrcEn-CN1UrxZXH0un
https://colab.research.google.com/drive/1HrTj5j9vo8IovHVnZHfcvAju3ZD1_X5f
https://colab.research.google.com/drive/13jkx5ax73hJsJnftUHLtePdaUX_j3U4X
https://colab.research.google.com/drive/1iNW7bZebJ7PpCFfKJPLLsFykHCxAPIjV
https://colab.research.google.com/drive/1EMLMitTHcxD1odQStMhqpgu1LlfDDThH
https://colab.research.google.com/drive/1e16rdl2z3ZI8wCUnuG2HGOHhO6RB0Nvf
https://colab.research.google.com/drive/1Fn4N12W9AV6ws6W7vxNHIZtB_M2N5lGI
https://colab.research.google.com/drive/1ewhSOoefu6qaq6a1gU6Af5S2oWblkFgw
https://colab.research.google.com/drive/1HAY84b9KOHevCDo8YP9AtYfEHZo__Tjq
https://colab.research.google.com/drive/1kDxZ6M—q_5DJL_Cej_gZXv52lMlIi35
https://colab.research.google.com/drive/1kDK8LNS6SA9ir5wUug2UGAPTZXseN8p6
https://colab.research.google.com/drive/1-JHkUAaH6rQ6_kOY7t5iBg3js9eHhJ8O
https://colab.research.google.com/drive/19AfdLeWhBUrq5Ozv4WBqe4sFVpCg0O6S
https://colab.research.google.com/drive/14kCiX_P2RrPmHgwYBJQNU14B-CitLfej
https://colab.research.google.com/drive/1xUUyA3i96vXbiwRfbTVblMJHYtVKVKP0
https://colab.research.google.com/drive/1yT6lozNv0FBfeq_33czBmMaDvJCfQ3pH
https://colab.research.google.com/drive/1CCu1MQHLIic2gkoo6OQjyaemsaK66gRx
https://colab.research.google.com/drive/1gpaRhxSTyZMbLyZqVQpXVgTM_2VMGY16
https://colab.research.google.com/drive/100YRXn6YgihmunTZ-L5MCOKoSJ2zmBA5
https://colab.research.google.com/drive/1z6i4ia6rh4doKbv06DYDkKtgEjeh-EkZ
https://colab.research.google.com/drive/1N1MimKAuNbevRE66XZML7Hq66y39o3m1
https://colab.research.google.com/drive/1GGCEJJdnIpq6udvqSVwB4NdSLCMyArNh
https://colab.research.google.com/drive/1kol1eDhk0x757Uu4Xi6vJ66rZl0fTrNC
https://colab.research.google.com/drive/1KuJ_JpIgvmeJalfsq5baEUXaqS6QU4Ad
https://colab.research.google.com/drive/1rL_9RNPIUjBusMivLizfCviQYzj934Ip
https://colab.research.google.com/drive/105qftPoYucY2ewhsO2T_43tCAJY2t3bz
https://colab.research.google.com/drive/1IKecs52IeKcWF_LO2jon1j_ZH2isc4oK
https://colab.research.google.com/drive/1JvGNwYws69yO17KB54645m1RcmDVysmB
https://colab.research.google.com/drive/18iSsubGyGhnKQ7k7tZ4KcmrX8wP3vx6o
https://colab.research.google.com/drive/1iPCIgMYoXExHECIwnErWs0isDfYKBiQb
https://colab.research.google.com/drive/1Od8EqMXitJEHXdBGXOBaLlCBOLHlfnbV
https://colab.research.google.com/drive/1I1iC-eWWso5qmosKpImsOoAxdflUmfl8
https://colab.research.google.com/drive/1vcfGEV6Vr4Gs9sZbHnrbSduhVeHSPIEx
https://colab.research.google.com/drive/1Fs-p691jWPClfo712m7q_rCLeW7R1kfS
https://colab.research.google.com/drive/1ChMuO3cVeMarNCV1fkXW18OHDuZkcltK
https://colab.research.google.com/drive/1bwS_96Zp8YTx_B8hjIHohOceTdV5wye4
https://colab.research.google.com/drive/1eqbBEWQ1m4feCuOpnP1UdyBZ3UQ4m1Tu
https://colab.research.google.com/drive/1FGG2uFj4RhPEAnHKFC6OpPzooaBdd7uO
https://colab.research.google.com/drive/1X9bFOLjjgIFn6sJVemzikEsLirDfjmtl
https://colab.research.google.com/drive/1y204YuKj6H5nXjvUkOuiaeoiH_LcNFyJ
https://colab.research.google.com/drive/141ZJSeflQB0Mgj0ZwewerJlRlPEdIelk
https://colab.research.google.com/drive/163nPfXLKHTjcJ2tNd_VeUg193i6e48j8
https://colab.research.google.com/drive/16Rppdp251mJaJNYve2rUfvM2Jv8iq0ty
https://colab.research.google.com/drive/1VB2DkdfMJ9uAu2dDLn5KPLlC9CKnGA23
https://colab.research.google.com/drive/1r3-TrsUWzc8zJS5s5CiKMBgybRZQQ_Au
https://colab.research.google.com/drive/1Qnvx80kN-pTpKuan_buRW5hLQ_JMWSt8
https://colab.research.google.com/drive/1NFqPnZ8rXy_mY9acuf7IcKMm0C4G_NsA
https://colab.research.google.com/drive/1QM_cQFBuwGue6gQm6mZmXug5uG_1g8bB
https://colab.research.google.com/drive/1OD4uMxqZzUEg9Ip2ah8UQBfy7hSen762
https://colab.research.google.com/drive/1wz6xWAXBV0HzWI-tgWz-v-8hsPWZ5iSq
https://colab.research.google.com/drive/1XVdn6xEuyzUyQ9GvRtFjLK_CvNKIGMHR
https://colab.research.google.com/drive/1zT5y7-Ve_JWzvEuZd-pmPAUtGmmIo077
https://colab.research.google.com/drive/1c1WNm4x8y5B8yPni97MSeapzJtAxuSaW
https://colab.research.google.com/drive/1ODFc8hwSkkoaDTf7ql5tRuzjNMck9NiL
https://colab.research.google.com/drive/1ZnxS3VUc11gjzjClZ2jgCgh_fgnsHMcf
https://colab.research.google.com/drive/1ovqo4ItXlIS5q7gKw1QPPAO-xidcloYF
https://colab.research.google.com/drive/1sLYUiHJLKQmchCeWKc9fkCb_GCfyffyV
https://colab.research.google.com/drive/1kdQTYarXtQm6rlKHTPhdiAw_78aG6iHW
https://colab.research.google.com/drive/1whckEx_jnbmbD5Za8eibImdMAMoGJk6P
https://colab.research.google.com/drive/1je8ajjeun1rFTwkXaTvb2rbrwBHv3N1-
https://colab.research.google.com/drive/1zdMEV5xPTq2B-jMqdTbaPSNyv6orYLbQ
https://colab.research.google.com/drive/1DX8F-L6BTTWYRJdRq-QcBRWO78Gy89G
https://colab.research.google.com/drive/1Syr-ILdE0bUHLJLX1QJnb0Odd0SYPZj
https://colab.research.google.com/drive/15eCF5fOpgzEdX4flS-oaUlryqiwU-cYo
https://colab.research.google.com/drive/1dG8QmCBkIoNbmxx0kFH8Np8Cmslb91V0
https://colab.research.google.com/drive/1I4bGVmKs6OzG88N0UVv7iXxXw8VGDV6p
https://colab.research.google.com/drive/12-7b4FbxYvDMGpgH162tdgFPbfRR7aPW
https://colab.research.google.com/drive/18a8hxdmFrGjm_4XmchQXSf5ulRsl7_2o
https://colab.research.google.com/drive/1pb7OnjK9VVnqYTbscoYZAV0E055F5Ip2
https://colab.research.google.com/drive/19xrkGpGYgM3UYSa3B—9SlQxOZrbce9R
https://colab.research.google.com/drive/1NF0tLGFx4O4XPkDp3fzu3zjXxBqm9DFf
https://colab.research.google.com/drive/14pCfRH69gmHfy9Erx_DDXcqXDM4ySHyS
https://colab.research.google.com/drive/1upigYqbzSeY8InEb470JuhhVu-9QO9Zl
https://colab.research.google.com/drive/1Tm7DeMyYrQF-YAig2avo_qOPWWo_wZZo
https://colab.research.google.com/drive/1ZfxP3rDviZ-fuMj-XItZcnHKS-e5PM8F
https://colab.research.google.com/drive/1oCnkTz5a_aZFpMqIndyQ7YSF6M3koqm5
https://colab.research.google.com/drive/1EtFH3HHlz5otYLOVTd_2OTZzW2Cc5NWk
https://colab.research.google.com/drive/1EoQB-L8-xFsA3TUcPNbUsBK311zmob6w
https://colab.research.google.com/drive/15E_CieZfI666avic_hItMduyV562w2lh
https://colab.research.google.com/drive/1nGruX9VCkjE955x01eB8iJe-MlK8gM54
https://colab.research.google.com/drive/1xcEPlxtvG_rbUKO83F-afNUM4g4GNmS2
https://colab.research.google.com/drive/1N6EvQC-o-5WOatqdp7Mrev2XzpsB8Dd1
https://colab.research.google.com/drive/1GOgwJqa5G8d4xCXAN8F3tRnln2pHU4V0
https://colab.research.google.com/drive/1h_prXY7nElA98xCB5otxZmLSemWBMfJH
https://colab.research.google.com/drive/1WaHTObnBe843CNJO24wJ0k5pASUbPNBO
https://colab.research.google.com/drive/1j6f_Ykpg0n5B5vJq4mnU8gWXR3UYroia
https://colab.research.google.com/drive/1vWk5n8WrNjNuMkw2X0b21XotE_TCIeYw
https://colab.research.google.com/drive/1G3trAGw-XN9lxzfriWDerZhrHz3C_xdk
https://colab.research.google.com/drive/1-RFbIcvziNeL9H-tFHp9ENBLh5Zlnf7F
https://colab.research.google.com/drive/1dzqU9L4ZaMsqtyG58Xv6igSSN1t8aZco
https://colab.research.google.com/drive/1HFPZICzmxUuxMtbgHVit0C6tV_JF6Zr0
https://colab.research.google.com/drive/1Dmyq-8tHf5RGYCIF8MjRUh3XRq0ulyu9
https://colab.research.google.com/drive/1mxuZzmFb9fvAZlVCh3b5LTZt9V2dq3I_
https://colab.research.google.com/drive/1f7gkN-n8tFjZ4tfFFRwqDR518zxpdIM9
https://colab.research.google.com/drive/1iUk4qug4QYtOL_aKuBQb8GoWQ1DLxCbW
https://colab.research.google.com/drive/1LWRZeszFIy-6SE4eFS9kNdQYoYHKWtEn
https://colab.research.google.com/drive/1jYqVEmuDN68uWG3bGWVfOUfJgNbOPh0G
https://colab.research.google.com/drive/1ohwPOlwwr7fN8XwWWLoeUKzY3BhwE955
https://colab.research.google.com/drive/1vmANVD0dJGXNBXphGfZuLmrz3OqH2YST
https://colab.research.google.com/drive/1ZzidQpxb2jY_tCB9auGhsXn0Hc_5WOVT
https://colab.research.google.com/drive/1bGFFOv5EbQ6u9ikvZvX30InDygDUvofy
https://colab.research.google.com/drive/18p3tlWHeroJavukpdqSLP0hytOl702pZ
https://colab.research.google.com/drive/1mEx_GVTtlextYne2NqmzQspMHQ0Y7xLe
https://colab.research.google.com/drive/1VRUEst37oJBWiGljMPXKKvDk71rNhrOy
https://colab.research.google.com/drive/1vhZ_s9_OcK2_vOZLi3BBNAaSwEN8J4iO
https://colab.research.google.com/drive/1APgIks1pl-1Z5J9jjvhRZ9FTRWE_5D2J
https://colab.research.google.com/drive/1kOdbCrscJGPsBMfWdJ2qMpNEcJT2LFAD
https://colab.research.google.com/drive/1HfShgAKOVqIz8hP0zDrBOY9QlTXo_u2-
https://colab.research.google.com/drive/143yfvdfyr7HQHS2bnPhsNCH2HB27pSwc
https://colab.research.google.com/drive/1SdGlNbVDykd8dHVpmqL6RRcIEzj2h4M9
https://colab.research.google.com/drive/1-6UAaE0wa6_el8JLKx-C82fVGRmn2qdJ
https://colab.research.google.com/drive/1lAOorkrEultRr_mNPxm3lTmw25x5VwbU
https://colab.research.google.com/drive/1kYdqsLhP_ufbcn7vddirvt385jOq66oa
https://colab.research.google.com/drive/1c_cZj2ogZxbOlqY_EJ-9-icuuz8yJUWh
https://colab.research.google.com/drive/14ExO6COkOi0yuuHHlPxIoD7GytzD5xyX
https://colab.research.google.com/drive/1FFCNXe3dz-gm6bA5L_wVhmqh8JmTovKR
https://colab.research.google.com/drive/1tMfrVC7TFUfl5Sk8M7gmpNK90-qOP_hC
https://colab.research.google.com/drive/1GMKVdt97fZDDkaTevJRFjOIm90rRPryz
https://colab.research.google.com/drive/1QVMLYjdJ6DXdOzmBZx6sFtVCJ4E1e3rR
https://colab.research.google.com/drive/18TzSXzE5k1LucszFNcpegT34YgwBiK_U
https://colab.research.google.com/drive/1WXGP7_jaRgupuZUw1zsXrVNAQL5GaExw
https://colab.research.google.com/drive/1xP7-D--8bVVJ1S0TzWXAKe-4qrFFKlY
https://colab.research.google.com/drive/1Z9amJBV7A1dXfF8U-ChOcrDm6zZxEWMW
https://colab.research.google.com/drive/1qrYVb7TFF9KNw4PQmKXyGTrv-V6fwb_Z
https://colab.research.google.com/drive/1JgN4LlJOyEISfihl5ae531eMICDRrEs0
https://colab.research.google.com/drive/1wTzDksnwfVapnHrnjcnNsnPEHOsWsAvo
https://colab.research.google.com/drive/1IIwW-nINlzL9_pPXDIqebQVzL19xq-dS
https://colab.research.google.com/drive/1uy7vy-GGsi7HpYk5QafNpABq8bRtOryq
https://colab.research.google.com/drive/1_k32niAXp0VyXcbo1vJnIglJs9mP2spl
https://colab.research.google.com/drive/1eGqqSz086EzHCegg6Gw8FXgD0zWfWVD4
https://colab.research.google.com/drive/11nCPJ3KVHMzIMGYHcHEdJx70Yt0OCoaG
https://colab.research.google.com/drive/1MuCqoRQal1vPcup8qiZOns_756ZjH2JD
https://colab.research.google.com/drive/1M2ylFSfOAsXbxyl2guaZECVpmsoJoo6p
https://colab.research.google.com/drive/13KpA16gu-my8SM3mBB0WBBLbuza212nL
https://colab.research.google.com/drive/1rXagWhmZvYYT1Dy9gDH8B0gO_UXkRyII
https://colab.research.google.com/drive/1I2EsmHqm_Z9HXvDiqYVXn2J-b0LVWG3V
https://colab.research.google.com/drive/1XA1pFG-FxYDtOo1uhTpEw5n1NCSKmLDC
https://colab.research.google.com/drive/1-KKleX8EZ8NB-3_1eMMxwvNEXMLW1ips
https://colab.research.google.com/drive/1rvqelOQ5jiWpgWDKtBTv0STyzgCpkZ3V
https://colab.research.google.com/drive/1YcLY5vqsBcs9BkV08OoBc7akYmYNjqTV
https://colab.research.google.com/drive/1bDe11nDOBMHofyPRIH2uRHJIHhCBc3z
https://colab.research.google.com/drive/1izmOUmvbK866RsbUqMehk9Qdvetktp26
https://colab.research.google.com/drive/1qluJN95bteNxJ-K7VzxtNFL2CclZvOX7
https://colab.research.google.com/drive/1qwsnc5fubE87UKIq8WkdLMz9tIWzt4dz
https://colab.research.google.com/drive/1u_BKoTf-RA0-ai_L6octA4HgMFK0-9v7
https://colab.research.google.com/drive/11KS74Dy9kC7ZwyP9Imw3Pe78uRW67Qmj
https://colab.research.google.com/drive/1OrYK8mu9dKPYmcmn0z7Zw-Sf3TjaQZvt
https://colab.research.google.com/drive/1O-I82JvY73GQg2F1XdIwKTIpDzMicMQW
https://colab.research.google.com/drive/1UDMfW4RD-nujz05zh83DObgjPHJaqJZL
https://colab.research.google.com/drive/1r6RoksL-gia2_m7r5tAD2-P5wgn5aIGL
https://colab.research.google.com/drive/1VE2XFJdpy1frpVYyrtBu7GQMz5zycg0q
https://colab.research.google.com/drive/1RMvWpxyj8mFR7498YcuEb5MVKR3i0LEt
https://colab.research.google.com/drive/1S714keRsEdReYYfudVVAYvGqtHVHEyab
https://colab.research.google.com/drive/1suixTpl8vktJYyMD7jUscobDyMPeTVE3
https://colab.research.google.com/drive/1VuzttkGFzEC76LmFWWsHLb9rSA9xXgs3
https://colab.research.google.com/drive/1YMfjAsnf6r9sDqp6qCKDIdBssBAk—7x
https://colab.research.google.com/drive/17TCuNdgdNSHuSB-gcMlcgbWjELWqVu4G
https://colab.research.google.com/drive/1_x055AhcdhXqIt0anl2l6Mozl4trB94F
https://colab.research.google.com/drive/1jEL_aACJTL6TmuB5TJUzMz2xYxM8pGmY
https://colab.research.google.com/drive/15PAm-HmKMfCm_yXGamJ2uIu5imOHQSfC
https://colab.research.google.com/drive/1LZ0rV5P9s46_7wkje7gv7law-Zf-K6xx
https://colab.research.google.com/drive/14NeVeLA33ovWG0jd_4nmhb7iEwvWP6xW
https://colab.research.google.com/drive/1cOSwS51t5M4kRBg_MZvUFMNUA3tXYBrU
https://colab.research.google.com/drive/1SOfT1siin9jV_BD4-o31tE08XbOWslX8
https://colab.research.google.com/drive/1VbS_dw_ebYXUpm687MMFxnA1coGFA-Tt
https://colab.research.google.com/drive/1xrOfd7mfL9DjxrQjyRCH9M2sIf8vlLqK
https://colab.research.google.com/drive/1nA5Y6eXW4Pz3F3QQSYWqm4fnHlrvLtzO
https://colab.research.google.com/drive/1QvTyogv64OF5wKOICwUC87ulX4avqNd4
https://colab.research.google.com/drive/1bt06stoP80wtUKjpoHAcoZ40zNwzA_K3
https://colab.research.google.com/drive/1hyCmo56hRIZ_keD578rTity6oKLF89cK
https://colab.research.google.com/drive/1MG7SyDw80PWQA7bVA184LZWrY3RCE0ou
https://colab.research.google.com/drive/1Eo8Oq3IQ8YLGq-JRbipn2tQ03N17nL9N
https://colab.research.google.com/drive/1LtNAmMhUQCttpeIU2NUFdZkEh56Mwi3i
https://colab.research.google.com/drive/1vsmq4w9mkuK6PtUIdKk_U9OI5dS4EGY3
https://colab.research.google.com/drive/1r-DS12L5YECABlk9fB02yDL9pp6dLjFL
https://colab.research.google.com/drive/1LJpmPnBxZg-m8ROWlxVeBMmRnx1popo1
https://colab.research.google.com/drive/1FKhTGL2Qr7raAJW3THaP0-X-Op02Kbev
---
license: bigscience-openrail-m
datasets:
- databricks/databricks-dolly-15k
metrics:
- brier_score
pipeline_tag: zero-shot-classification
tags:
- art
---
|
sofia-todeschini/PubMedBERT-Large-LitCovid-v1.0
|
sofia-todeschini
| 2023-06-12T17:06:14Z | 107 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"bert",
"text-classification",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2023-06-12T15:33:21Z |
---
license: mit
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: PubMedBERT-Large-LitCovid-v1.0
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# PubMedBERT-Large-LitCovid-v1.0
This model is a fine-tuned version of [microsoft/BiomedNLP-PubMedBERT-large-uncased-abstract](https://huggingface.co/microsoft/BiomedNLP-PubMedBERT-large-uncased-abstract) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1044
- F1: 0.9021
- Roc Auc: 0.9367
- Accuracy: 0.8006
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
### Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
| 0.1103 | 1.0 | 6240 | 0.1044 | 0.9021 | 0.9367 | 0.8006 |
### Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
|
huggingtweets/lillygvtuber
|
huggingtweets
| 2023-06-12T17:01:58Z | 132 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"huggingtweets",
"en",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T17:01:48Z |
---
language: en
thumbnail: https://github.com/borisdayma/huggingtweets/blob/master/img/logo.png?raw=true
tags:
- huggingtweets
widget:
- text: "My dream is"
---
<div class="inline-flex flex-col" style="line-height: 1.5;">
<div class="flex">
<div
style="display:inherit; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('https://pbs.twimg.com/profile_images/1597529813090172928/a2KDahqS_400x400.jpg')">
</div>
<div
style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')">
</div>
<div
style="display:none; margin-left: 4px; margin-right: 4px; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url('')">
</div>
</div>
<div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 AI BOT 🤖</div>
<div style="text-align: center; font-size: 16px; font-weight: 800">Lilly Gracie</div>
<div style="text-align: center; font-size: 14px;">@lillygvtuber</div>
</div>
I was made with [huggingtweets](https://github.com/borisdayma/huggingtweets).
Create your own bot based on your favorite user with [the demo](https://colab.research.google.com/github/borisdayma/huggingtweets/blob/master/huggingtweets-demo.ipynb)!
## How does it work?
The model uses the following pipeline.
To understand how the model was developed, check the [W&B report](https://wandb.ai/wandb/huggingtweets/reports/HuggingTweets-Train-a-Model-to-Generate-Tweets--VmlldzoxMTY5MjI).
## Training data
The model was trained on tweets from Lilly Gracie.
| Data | Lilly Gracie |
| --- | --- |
| Tweets downloaded | 1576 |
| Retweets | 89 |
| Short tweets | 208 |
| Tweets kept | 1279 |
[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/0um51eml/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
## Training procedure
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @lillygvtuber's tweets.
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/94aijvbx) for full transparency and reproducibility.
At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/94aijvbx/artifacts) is logged and versioned.
## How to use
You can use this model directly with a pipeline for text generation:
```python
from transformers import pipeline
generator = pipeline('text-generation',
model='huggingtweets/lillygvtuber')
generator("My dream is", num_return_sequences=5)
```
## Limitations and bias
The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
In addition, the data present in the user's tweets further affects the text generated by the model.
## About
*Built by Boris Dayma*
[](https://twitter.com/intent/follow?screen_name=borisdayma)
For more details, visit the project repository.
[](https://github.com/borisdayma/huggingtweets)
|
Shrey-1329/ppo-Huggy
|
Shrey-1329
| 2023-06-12T17:01:52Z | 12 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"Huggy",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-Huggy",
"region:us"
] |
reinforcement-learning
| 2023-06-12T17:01:47Z |
---
library_name: ml-agents
tags:
- Huggy
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-Huggy
---
# **ppo** Agent playing **Huggy**
This is a trained model of a **ppo** agent playing **Huggy**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: Shrey-1329/ppo-Huggy
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
Jeibros/ppo-SnowballTarget
|
Jeibros
| 2023-06-12T16:53:28Z | 11 | 0 |
ml-agents
|
[
"ml-agents",
"tensorboard",
"onnx",
"SnowballTarget",
"deep-reinforcement-learning",
"reinforcement-learning",
"ML-Agents-SnowballTarget",
"region:us"
] |
reinforcement-learning
| 2023-06-06T07:27:02Z |
---
library_name: ml-agents
tags:
- SnowballTarget
- deep-reinforcement-learning
- reinforcement-learning
- ML-Agents-SnowballTarget
---
# **ppo** Agent playing **SnowballTarget**
This is a trained model of a **ppo** agent playing **SnowballTarget**
using the [Unity ML-Agents Library](https://github.com/Unity-Technologies/ml-agents).
## Usage (with ML-Agents)
The Documentation: https://unity-technologies.github.io/ml-agents/ML-Agents-Toolkit-Documentation/
We wrote a complete tutorial to learn to train your first agent using ML-Agents and publish it to the Hub:
- A *short tutorial* where you teach Huggy the Dog 🐶 to fetch the stick and then play with him directly in your
browser: https://huggingface.co/learn/deep-rl-course/unitbonus1/introduction
- A *longer tutorial* to understand how works ML-Agents:
https://huggingface.co/learn/deep-rl-course/unit5/introduction
### Resume the training
```bash
mlagents-learn <your_configuration_file_path.yaml> --run-id=<run_id> --resume
```
### Watch your Agent play
You can watch your agent **playing directly in your browser**
1. If the environment is part of ML-Agents official environments, go to https://huggingface.co/unity
2. Step 1: Find your model_id: Jeibros/ppo-SnowballTarget
3. Step 2: Select your *.nn /*.onnx file
4. Click on Watch the agent play 👀
|
ziq/llama-guanaco-7b-ingbetic
|
ziq
| 2023-06-12T16:37:47Z | 0 | 0 | null |
[
"tensorboard",
"generated_from_trainer",
"license:other",
"region:us"
] | null | 2023-06-12T15:28:31Z |
---
license: other
tags:
- generated_from_trainer
model-index:
- name: llama-guanaco-7b-ingbetic
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# llama-guanaco-7b-ingbetic
This model is a fine-tuned version of [decapoda-research/llama-7b-hf](https://huggingface.co/decapoda-research/llama-7b-hf) on the None dataset.
It achieves the following results on the evaluation set:
- eval_loss: 1.0322
- eval_runtime: 65.492
- eval_samples_per_second: 27.973
- eval_steps_per_second: 3.497
- epoch: 0.12
- step: 80
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- training_steps: 100
### Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.12.1.dev0
- Tokenizers 0.13.3
|
felipemaiapolo/legalnlp-bert
|
felipemaiapolo
| 2023-06-12T16:30:32Z | 413 | 6 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"fill-mask",
"pt",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
fill-mask
| 2023-06-12T15:43:10Z |
---
license: mit
language:
- pt
---
# BERTikal (aka `legalnlp-bert`)
BERTikal [1] is a cased BERT-base model for the Brazilian legal language and was trained from the BERTimbau's [2] checkpoint using Brazilian legal texts. More details on the datasets and training procedures can be found in [1].
**Please check Legal-NLP** out for more resources on (PT-BR) legal natural language processing (https://github.com/felipemaiapolo/legalnlp).
**Please cite as** Polo, Felipe Maia, et al. "LegalNLP-Natural Language Processing methods for the Brazilian Legal Language." Anais do XVIII Encontro Nacional de Inteligência Artificial e Computacional. SBC, 2021.
@inproceedings{polo2021legalnlp,
title={LegalNLP-Natural Language Processing methods for the Brazilian Legal Language},
author={Polo, Felipe Maia and Mendon{\c{c}}a, Gabriel Caiaffa Floriano and Parreira, Kau{\^e} Capellato J and Gianvechio, Lucka and Cordeiro, Peterson and Ferreira, Jonathan Batista and de Lima, Leticia Maria Paz and do Amaral Maia, Ant{\^o}nio Carlos and Vicente, Renato},
booktitle={Anais do XVIII Encontro Nacional de Intelig{\^e}ncia Artificial e Computacional},
pages={763--774},
year={2021},
organization={SBC}
}
## Usage
### Ex. Loading model for general use
```python
from transformers import AutoTokenizer # Or BertTokenizer
from transformers import AutoModelForPreTraining # Or BertForPreTraining for loading pretraining heads
from transformers import AutoModel # or BertModel, for BERT without pretraining heads
model = AutoModelForPreTraining.from_pretrained('felipemaiapolo/legalnlp-bert')
tokenizer = AutoTokenizer.from_pretrained('felipemaiapolo/legalnlp-bert', do_lower_case=False)
```
### Ex. BERT embeddings
```python
from transformers import pipeline
pipe = pipeline("feature-extraction", model='felipemaiapolo/legalnlp-bert')
encoded_sentence = pipe('Juíz negou o recurso.')
```
### Ex. Masked language modeling prediction
```python
from transformers import pipeline
pipe = pipeline('fill-mask', model='felipemaiapolo/legalnlp-bert')
pipe('Juíz negou o [MASK].')
# [{'score': 0.6387444734573364,
# 'token': 7608,
# 'token_str': 'julgamento',
# 'sequence': 'juiz negou o julgamento.'},
# {'score': 0.09632532298564911,
# 'token': 7509,
# 'token_str': 'voto',
# 'sequence': 'juiz negou o voto.'},
# {'score': 0.06424401700496674,
# 'token': 17225,
# 'token_str': 'julgado',
# 'sequence': 'juiz negou o julgado.'},
# {'score': 0.05929475650191307,
# 'token': 8190,
# 'token_str': 'recurso',
# 'sequence': 'juiz negou o recurso.'},
# {'score': 0.011442390270531178,
# 'token': 6330,
# 'token_str': 'registro',
# 'sequence': 'juiz negou o registro.'}]
```
# References
[1] Polo, Felipe Maia, et al. "LegalNLP-Natural Language Processing methods for the Brazilian Legal Language." Anais do XVIII Encontro Nacional de Inteligência Artificial e Computacional. SBC, 2021.
[2] Souza, F., Nogueira, R., and Lotufo, R. (2020). BERTimbau: pretrained BERT
models for Brazilian Portuguese. In 9th Brazilian Conference on Intelligent
Systems, BRACIS, Rio Grande do Sul, Brazil, October 20-23
|
rizquuula/IndoBERT-SQuADv2-16-2e-06-0.01-5
|
rizquuula
| 2023-06-12T16:17:16Z | 116 | 0 |
transformers
|
[
"transformers",
"pytorch",
"bert",
"question-answering",
"generated_from_trainer",
"license:mit",
"endpoints_compatible",
"region:us"
] |
question-answering
| 2023-06-12T08:50:27Z |
---
license: mit
tags:
- generated_from_trainer
model-index:
- name: IndoBERT-SQuADv2_1686559661-16-2e-06-0.01-5
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# IndoBERT-SQuADv2_1686559661-16-2e-06-0.01-5
This model is a fine-tuned version of [indobenchmark/indobert-base-p1](https://huggingface.co/indobenchmark/indobert-base-p1) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.2709
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:-----:|:---------------:|
| 2.8367 | 1.0 | 8145 | 2.4799 |
| 2.2767 | 2.0 | 16290 | 2.3464 |
| 2.12 | 3.0 | 24435 | 2.3053 |
| 2.0313 | 4.0 | 32580 | 2.2777 |
| 1.9826 | 5.0 | 40725 | 2.2709 |
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3
|
mzbac/tulu-grammar-13b
|
mzbac
| 2023-06-12T16:01:57Z | 7 | 2 |
transformers
|
[
"transformers",
"llama",
"text-generation",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T10:53:19Z |
A fine-tuned LLaMa model for grammar correction, fine-tuned on a dataset generated via ChatGPT for the software development context.
prompt format:
```
An AI tool that corrects user text grammar errors delimited by triple backticks to standard English.
INPUT: Correct the following [text=```{user input}```]
OUTPUT:
```
|
hopkins/strict-small-3g
|
hopkins
| 2023-06-12T16:00:42Z | 132 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"generated_from_trainer",
"dataset:generator",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T15:27:29Z |
---
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: strict-small-3g
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# strict-small-3g
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 512
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
### Training results
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3
|
hopkins/strict-small-3f
|
hopkins
| 2023-06-12T15:55:43Z | 132 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"generated_from_trainer",
"dataset:generator",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T15:22:34Z |
---
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: strict-small-3f
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# strict-small-3f
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 512
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
### Training results
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3
|
hopkins/strict-small-3e
|
hopkins
| 2023-06-12T15:27:14Z | 130 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"generated_from_trainer",
"dataset:generator",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T14:52:43Z |
---
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: strict-small-3e
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# strict-small-3e
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 512
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
### Training results
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3
|
Deojaklah/Memeeee
|
Deojaklah
| 2023-06-12T15:14:00Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T15:12:08Z |
---
license: creativeml-openrail-m
---
|
wootwoot/abyssorangemix3-fp16
|
wootwoot
| 2023-06-12T14:54:59Z | 34 | 2 |
diffusers
|
[
"diffusers",
"safetensors",
"text-to-image",
"en",
"license:creativeml-openrail-m",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us"
] |
text-to-image
| 2023-06-12T14:42:20Z |
---
license: creativeml-openrail-m
language:
- en
library_name: diffusers
pipeline_tag: text-to-image
---
### Based off [WarriorMama777/OrangeMixs](https://huggingface.co/WarriorMama777/OrangeMixs)
All credits go to the original author and all the author of AbyssOrangeMix3's ancestor models
### Diffusers
The original AbyssOrangeMix3 model converted to be used with the [🧨Diffusers library](https://github.com/huggingface/diffusers)
|
yanmul011897/pep
|
yanmul011897
| 2023-06-12T14:44:37Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T14:33:45Z |
---
license: creativeml-openrail-m
---
|
cartesinus/slurp-intent_baseline-xlm_r-en
|
cartesinus
| 2023-06-12T14:44:18Z | 103 | 1 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"xlm-roberta",
"text-classification",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2023-04-27T08:55:37Z |
---
license: mit
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: slurp-intent_baseline-xlm_r-en
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# slurp-intent_baseline-xlm_r-en
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on an SLURP dataset.
It achieves the following results on the test set:
- Loss: 0.68222
- Accuracy: 0.8746
- F1: 0.8746
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 2.9687 | 1.0 | 720 | 1.3267 | 0.6955 | 0.6955 |
| 1.4534 | 2.0 | 1440 | 0.8053 | 0.8219 | 0.8219 |
| 0.6775 | 3.0 | 2160 | 0.6912 | 0.8421 | 0.8421 |
| 0.5624 | 4.0 | 2880 | 0.6377 | 0.8623 | 0.8623 |
| 0.3756 | 5.0 | 3600 | 0.6188 | 0.8746 | 0.8746 |
| 0.3346 | 6.0 | 4320 | 0.6548 | 0.8711 | 0.8711 |
| 0.2541 | 7.0 | 5040 | 0.6618 | 0.8751 | 0.8751 |
| 0.2243 | 8.0 | 5760 | 0.6662 | 0.8780 | 0.8780 |
| 0.212 | 9.0 | 6480 | 0.6673 | 0.8810 | 0.8810 |
| 0.1664 | 10.0 | 7200 | 0.6783 | 0.8810 | 0.8810 |
### Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
|
Brazilia/Feegrey
|
Brazilia
| 2023-06-12T14:32:56Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T14:30:04Z |
---
license: creativeml-openrail-m
---
|
SaturnMk2/ola
|
SaturnMk2
| 2023-06-12T14:27:58Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T14:27:23Z |
---
license: creativeml-openrail-m
---
|
hopkins/strict-small-3a
|
hopkins
| 2023-06-12T14:26:37Z | 130 | 0 |
transformers
|
[
"transformers",
"pytorch",
"gpt2",
"text-generation",
"generated_from_trainer",
"dataset:generator",
"license:mit",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2023-06-12T13:34:08Z |
---
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: strict-small-3a
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# strict-small-3a
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 1024
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
### Training results
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3
|
nolanaatama/xxmx_9rlstc
|
nolanaatama
| 2023-06-12T14:04:32Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T12:34:43Z |
---
license: creativeml-openrail-m
---
|
sampathlonka/q-FrozenLake-v1-4x4-noSlippery
|
sampathlonka
| 2023-06-12T13:50:37Z | 0 | 0 | null |
[
"FrozenLake-v1-4x4-no_slippery",
"q-learning",
"reinforcement-learning",
"custom-implementation",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T13:29:11Z |
---
tags:
- FrozenLake-v1-4x4-no_slippery
- q-learning
- reinforcement-learning
- custom-implementation
model-index:
- name: q-FrozenLake-v1-4x4-noSlippery
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: FrozenLake-v1-4x4-no_slippery
type: FrozenLake-v1-4x4-no_slippery
metrics:
- type: mean_reward
value: 1.00 +/- 0.00
name: mean_reward
verified: false
---
# **Q-Learning** Agent playing1 **FrozenLake-v1**
This is a trained model of a **Q-Learning** agent playing **FrozenLake-v1** .
## Usage
```python
model = load_from_hub(repo_id="sampathlonka/q-FrozenLake-v1-4x4-noSlippery", filename="q-learning.pkl")
# Don't forget to check if you need to add additional attributes (is_slippery=False etc)
env = gym.make(model["env_id"])
```
|
hannahh7/Reinforce-pixelcopter
|
hannahh7
| 2023-06-12T13:40:35Z | 0 | 0 | null |
[
"Pixelcopter-PLE-v0",
"reinforce",
"reinforcement-learning",
"custom-implementation",
"deep-rl-class",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-05-21T12:54:36Z |
---
tags:
- Pixelcopter-PLE-v0
- reinforce
- reinforcement-learning
- custom-implementation
- deep-rl-class
model-index:
- name: Reinforce-pixelcopter
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Pixelcopter-PLE-v0
type: Pixelcopter-PLE-v0
metrics:
- type: mean_reward
value: 15.90 +/- 10.00
name: mean_reward
verified: false
---
# **Reinforce** Agent playing **Pixelcopter-PLE-v0**
This is a trained model of a **Reinforce** agent playing **Pixelcopter-PLE-v0** .
To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction
|
achillon/ppo-LunarLander-v2
|
achillon
| 2023-06-12T13:14:47Z | 0 | 0 |
stable-baselines3
|
[
"stable-baselines3",
"LunarLander-v2",
"deep-reinforcement-learning",
"reinforcement-learning",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T09:41:39Z |
---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: PPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLander-v2
type: LunarLander-v2
metrics:
- type: mean_reward
value: 232.77 +/- 23.28
name: mean_reward
verified: false
---
# **PPO** Agent playing **LunarLander-v2**
This is a trained model of a **PPO** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
## Usage (with Stable-baselines3)
TODO: Add your code
```python
from stable_baselines3 import ...
from huggingface_sb3 import load_from_hub
...
```
|
rainymyself/q-FrozenLake-v1-4x4-noSlippery
|
rainymyself
| 2023-06-12T13:07:46Z | 0 | 0 | null |
[
"FrozenLake-v1-4x4-no_slippery",
"q-learning",
"reinforcement-learning",
"custom-implementation",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T13:07:44Z |
---
tags:
- FrozenLake-v1-4x4-no_slippery
- q-learning
- reinforcement-learning
- custom-implementation
model-index:
- name: q-FrozenLake-v1-4x4-noSlippery
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: FrozenLake-v1-4x4-no_slippery
type: FrozenLake-v1-4x4-no_slippery
metrics:
- type: mean_reward
value: 1.00 +/- 0.00
name: mean_reward
verified: false
---
# **Q-Learning** Agent playing1 **FrozenLake-v1**
This is a trained model of a **Q-Learning** agent playing **FrozenLake-v1** .
## Usage
```python
model = load_from_hub(repo_id="rainymyself/q-FrozenLake-v1-4x4-noSlippery", filename="q-learning.pkl")
# Don't forget to check if you need to add additional attributes (is_slippery=False etc)
env = gym.make(model["env_id"])
```
|
casque/clothingadjuster.jmXB
|
casque
| 2023-06-12T13:00:14Z | 0 | 1 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T12:59:31Z |
---
license: creativeml-openrail-m
---
|
Bly/CartPole-v1
|
Bly
| 2023-06-12T12:32:14Z | 0 | 0 | null |
[
"CartPole-v1",
"reinforce",
"reinforcement-learning",
"custom-implementation",
"deep-rl-class",
"model-index",
"region:us"
] |
reinforcement-learning
| 2023-06-12T12:32:05Z |
---
tags:
- CartPole-v1
- reinforce
- reinforcement-learning
- custom-implementation
- deep-rl-class
model-index:
- name: CartPole-v1
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: CartPole-v1
type: CartPole-v1
metrics:
- type: mean_reward
value: 485.20 +/- 44.40
name: mean_reward
verified: false
---
# **Reinforce** Agent playing **CartPole-v1**
This is a trained model of a **Reinforce** agent playing **CartPole-v1** .
To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction
|
lassl/bert-ko-small
|
lassl
| 2023-06-12T12:28:39Z | 130 | 2 |
transformers
|
[
"transformers",
"pytorch",
"safetensors",
"bert",
"pretraining",
"fill-mask",
"korean",
"lassl",
"ko",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
fill-mask
| 2022-03-02T23:29:05Z |
---
license: apache-2.0
language: ko
tags:
- fill-mask
- korean
- lassl
mask_token: "[MASK]"
widget:
- text: 대한민국의 수도는 [MASK] 입니다.
---
# LASSL bert-ko-small
## How to use
```python
from transformers import AutoModel, AutoTokenizer
model = AutoModel.from_pretrained("lassl/bert-ko-small")
tokenizer = AutoTokenizer.from_pretrained("lassl/bert-ko-small")
```
## Evaluation
Evaulation results will be released soon.
## Corpora
This model was trained from 702,437 examples (whose have 3,596,465,664 tokens). 702,437 examples are extracted from below corpora. If you want to get information for training, you should see `config.json`.
```bash
corpora/
├── [707M] kowiki_latest.txt
├── [ 26M] modu_dialogue_v1.2.txt
├── [1.3G] modu_news_v1.1.txt
├── [9.7G] modu_news_v2.0.txt
├── [ 15M] modu_np_v1.1.txt
├── [1008M] modu_spoken_v1.2.txt
├── [6.5G] modu_written_v1.0.txt
└── [413M] petition.txt
```
|
9wimu9/mt5-large-v7
|
9wimu9
| 2023-06-12T12:25:45Z | 105 | 0 |
transformers
|
[
"transformers",
"pytorch",
"mt5",
"text2text-generation",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2023-06-10T04:35:28Z |
{'exact_match': 56.173780487804876, 'f1': 79.52279646863565}
|
SaturnMk2/newarisa2
|
SaturnMk2
| 2023-06-12T12:23:27Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2023-06-12T12:22:53Z |
---
license: creativeml-openrail-m
---
|
Anis12/swin-tiny-patch4-window7-224-finetuned-eurosat
|
Anis12
| 2023-06-12T12:19:27Z | 213 | 0 |
transformers
|
[
"transformers",
"pytorch",
"tensorboard",
"swin",
"image-classification",
"generated_from_trainer",
"dataset:imagefolder",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
image-classification
| 2023-06-12T10:37:15Z |
---
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- imagefolder
metrics:
- accuracy
model-index:
- name: swin-tiny-patch4-window7-224-finetuned-eurosat
results:
- task:
name: Image Classification
type: image-classification
dataset:
name: imagefolder
type: imagefolder
config: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.5714285714285714
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# swin-tiny-patch4-window7-224-finetuned-eurosat
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0291
- Accuracy: 0.5714
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 90
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log | 1.0 | 1 | 0.9799 | 0.4286 |
| No log | 2.0 | 2 | 0.9703 | 0.4286 |
| No log | 3.0 | 3 | 0.9703 | 0.4286 |
| No log | 4.0 | 4 | 0.9699 | 0.4286 |
| No log | 5.0 | 5 | 0.9699 | 0.4286 |
| No log | 6.0 | 6 | 0.9881 | 0.4286 |
| No log | 7.0 | 7 | 0.9881 | 0.4286 |
| No log | 8.0 | 8 | 1.0213 | 0.4286 |
| No log | 9.0 | 9 | 1.0213 | 0.4286 |
| 0.426 | 10.0 | 10 | 1.0291 | 0.5714 |
| 0.426 | 11.0 | 11 | 1.0291 | 0.5714 |
| 0.426 | 12.0 | 12 | 0.9996 | 0.5714 |
| 0.426 | 13.0 | 13 | 0.9996 | 0.5714 |
| 0.426 | 14.0 | 14 | 0.8998 | 0.5714 |
| 0.426 | 15.0 | 15 | 0.8998 | 0.5714 |
| 0.426 | 16.0 | 16 | 0.8356 | 0.5714 |
| 0.426 | 17.0 | 17 | 0.8356 | 0.5714 |
| 0.426 | 18.0 | 18 | 0.8575 | 0.5714 |
| 0.426 | 19.0 | 19 | 0.8575 | 0.5714 |
| 0.324 | 20.0 | 20 | 0.9310 | 0.4286 |
| 0.324 | 21.0 | 21 | 0.9310 | 0.4286 |
| 0.324 | 22.0 | 22 | 1.0029 | 0.4286 |
| 0.324 | 23.0 | 23 | 1.0029 | 0.4286 |
| 0.324 | 24.0 | 24 | 1.0582 | 0.4286 |
| 0.324 | 25.0 | 25 | 1.0582 | 0.4286 |
| 0.324 | 26.0 | 26 | 1.0812 | 0.4286 |
| 0.324 | 27.0 | 27 | 1.0812 | 0.4286 |
| 0.324 | 28.0 | 28 | 1.0345 | 0.4286 |
| 0.324 | 29.0 | 29 | 1.0345 | 0.4286 |
| 0.2536 | 30.0 | 30 | 0.9996 | 0.4286 |
| 0.2536 | 31.0 | 31 | 0.9996 | 0.4286 |
| 0.2536 | 32.0 | 32 | 0.9401 | 0.5714 |
| 0.2536 | 33.0 | 33 | 0.9401 | 0.5714 |
| 0.2536 | 34.0 | 34 | 0.8978 | 0.5714 |
| 0.2536 | 35.0 | 35 | 0.8978 | 0.5714 |
| 0.2536 | 36.0 | 36 | 0.9056 | 0.5714 |
| 0.2536 | 37.0 | 37 | 0.9056 | 0.5714 |
| 0.2536 | 38.0 | 38 | 0.9364 | 0.5714 |
| 0.2536 | 39.0 | 39 | 0.9364 | 0.5714 |
| 0.2176 | 40.0 | 40 | 1.0523 | 0.5714 |
| 0.2176 | 41.0 | 41 | 1.0523 | 0.5714 |
| 0.2176 | 42.0 | 42 | 1.1687 | 0.4286 |
| 0.2176 | 43.0 | 43 | 1.1687 | 0.4286 |
| 0.2176 | 44.0 | 44 | 1.1968 | 0.4286 |
| 0.2176 | 45.0 | 45 | 1.1968 | 0.4286 |
| 0.2176 | 46.0 | 46 | 1.1604 | 0.4286 |
| 0.2176 | 47.0 | 47 | 1.1604 | 0.4286 |
| 0.2176 | 48.0 | 48 | 1.0505 | 0.4286 |
| 0.2176 | 49.0 | 49 | 1.0505 | 0.4286 |
| 0.1597 | 50.0 | 50 | 0.9059 | 0.5714 |
| 0.1597 | 51.0 | 51 | 0.9059 | 0.5714 |
| 0.1597 | 52.0 | 52 | 0.8606 | 0.5714 |
| 0.1597 | 53.0 | 53 | 0.8606 | 0.5714 |
| 0.1597 | 54.0 | 54 | 0.8946 | 0.5714 |
| 0.1597 | 55.0 | 55 | 0.8946 | 0.5714 |
| 0.1597 | 56.0 | 56 | 0.9643 | 0.5714 |
| 0.1597 | 57.0 | 57 | 0.9643 | 0.5714 |
| 0.1597 | 58.0 | 58 | 1.0598 | 0.5714 |
| 0.1597 | 59.0 | 59 | 1.0598 | 0.5714 |
| 0.1231 | 60.0 | 60 | 1.1833 | 0.5714 |
| 0.1231 | 61.0 | 61 | 1.1833 | 0.5714 |
| 0.1231 | 62.0 | 62 | 1.2730 | 0.5714 |
| 0.1231 | 63.0 | 63 | 1.2730 | 0.5714 |
| 0.1231 | 64.0 | 64 | 1.3132 | 0.4286 |
| 0.1231 | 65.0 | 65 | 1.3132 | 0.4286 |
| 0.1231 | 66.0 | 66 | 1.3025 | 0.4286 |
| 0.1231 | 67.0 | 67 | 1.3025 | 0.4286 |
| 0.1231 | 68.0 | 68 | 1.2702 | 0.4286 |
| 0.1231 | 69.0 | 69 | 1.2702 | 0.4286 |
| 0.1364 | 70.0 | 70 | 1.2411 | 0.4286 |
| 0.1364 | 71.0 | 71 | 1.2411 | 0.4286 |
| 0.1364 | 72.0 | 72 | 1.2222 | 0.4286 |
| 0.1364 | 73.0 | 73 | 1.2222 | 0.4286 |
| 0.1364 | 74.0 | 74 | 1.2257 | 0.4286 |
| 0.1364 | 75.0 | 75 | 1.2257 | 0.4286 |
| 0.1364 | 76.0 | 76 | 1.2552 | 0.4286 |
| 0.1364 | 77.0 | 77 | 1.2552 | 0.4286 |
| 0.1364 | 78.0 | 78 | 1.2701 | 0.5714 |
| 0.1364 | 79.0 | 79 | 1.2701 | 0.5714 |
| 0.0937 | 80.0 | 80 | 1.2753 | 0.5714 |
| 0.0937 | 81.0 | 81 | 1.2753 | 0.5714 |
| 0.0937 | 82.0 | 82 | 1.2797 | 0.5714 |
| 0.0937 | 83.0 | 83 | 1.2797 | 0.5714 |
| 0.0937 | 84.0 | 84 | 1.2840 | 0.5714 |
| 0.0937 | 85.0 | 85 | 1.2840 | 0.5714 |
| 0.0937 | 86.0 | 86 | 1.2895 | 0.5714 |
| 0.0937 | 87.0 | 87 | 1.2895 | 0.5714 |
| 0.0937 | 88.0 | 88 | 1.2917 | 0.5714 |
| 0.0937 | 89.0 | 89 | 1.2917 | 0.5714 |
| 0.1082 | 90.0 | 90 | 1.2931 | 0.5714 |
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
|
ABrinkmann/deberta-v3-large-finetuned-ner-10epochs
|
ABrinkmann
| 2023-06-12T12:19:22Z | 5 | 0 |
transformers
|
[
"transformers",
"pytorch",
"deberta-v2",
"token-classification",
"generated_from_trainer",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
token-classification
| 2023-06-12T10:55:08Z |
---
license: mit
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: deberta-v3-large-finetuned-ner-10epochs
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# deberta-v3-large-finetuned-ner-10epochs
This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0776
- Precision: 0.9247
- Recall: 0.9531
- F1: 0.9387
- Accuracy: 0.9890
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.0415 | 1.0 | 2261 | 0.0479 | 0.8894 | 0.9572 | 0.9220 | 0.9861 |
| 0.039 | 2.0 | 4522 | 0.0415 | 0.9051 | 0.9498 | 0.9269 | 0.9871 |
| 0.0304 | 3.0 | 6783 | 0.0390 | 0.9086 | 0.9579 | 0.9326 | 0.9883 |
| 0.0275 | 4.0 | 9044 | 0.0383 | 0.9146 | 0.9528 | 0.9333 | 0.9887 |
| 0.0224 | 5.0 | 11305 | 0.0421 | 0.9122 | 0.9583 | 0.9346 | 0.9884 |
| 0.0172 | 6.0 | 13566 | 0.0459 | 0.9204 | 0.9568 | 0.9382 | 0.9890 |
| 0.0122 | 7.0 | 15827 | 0.0561 | 0.9273 | 0.9429 | 0.9350 | 0.9885 |
| 0.0096 | 8.0 | 18088 | 0.0616 | 0.9223 | 0.9568 | 0.9393 | 0.9891 |
| 0.0051 | 9.0 | 20349 | 0.0718 | 0.9240 | 0.9568 | 0.9401 | 0.9890 |
| 0.0034 | 10.0 | 22610 | 0.0776 | 0.9247 | 0.9531 | 0.9387 | 0.9890 |
### Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3
|
Subsets and Splits
Filtered Qwen2.5 Distill Models
Identifies specific configurations of models by filtering cards that contain 'distill', 'qwen2.5', '7b' while excluding certain base models and incorrect model ID patterns, uncovering unique model variants.
Filtered Model Cards Count
Finds the count of entries with specific card details that include 'distill', 'qwen2.5', '7b' but exclude certain base models, revealing valuable insights about the dataset's content distribution.
Filtered Distill Qwen 7B Models
Filters for specific card entries containing 'distill', 'qwen', and '7b', excluding certain strings and patterns, to identify relevant model configurations.
Filtered Qwen-7b Model Cards
The query performs a detailed filtering based on specific keywords and excludes certain entries, which could be useful for identifying a specific subset of cards but does not provide deeper insights or trends.
Filtered Qwen 7B Model Cards
The query filters for specific terms related to "distilled" or "distill", "qwen", and "7b" in the 'card' column but excludes certain base models, providing a limited set of entries for further inspection.
Qwen 7B Distilled Models
The query provides a basic filtering of records to find specific card names that include keywords related to distilled Qwen 7b models, excluding a particular base model, which gives limited insight but helps in focusing on relevant entries.
Qwen 7B Distilled Model Cards
The query filters data based on specific keywords in the modelId and card fields, providing limited insight primarily useful for locating specific entries rather than revealing broad patterns or trends.
Qwen 7B Distilled Models
Finds all entries containing the terms 'distilled', 'qwen', and '7b' in a case-insensitive manner, providing a filtered set of records but without deeper analysis.
Distilled Qwen 7B Models
The query filters for specific model IDs containing 'distilled', 'qwen', and '7b', providing a basic retrieval of relevant entries but without deeper analysis or insight.
Filtered Model Cards with Distill Qwen2.
Filters and retrieves records containing specific keywords in the card description while excluding certain phrases, providing a basic count of relevant entries.
Filtered Model Cards with Distill Qwen 7
The query filters specific variations of card descriptions containing 'distill', 'qwen', and '7b' while excluding a particular base model, providing limited but specific data retrieval.
Distill Qwen 7B Model Cards
The query filters and retrieves rows where the 'card' column contains specific keywords ('distill', 'qwen', and '7b'), providing a basic filter result that can help in identifying specific entries.