|
|
--- |
|
|
license: apache-2.0 |
|
|
library_name: peft |
|
|
tags: |
|
|
- text-generation |
|
|
- alignment-handbook |
|
|
- generated_from_trainer |
|
|
- trl |
|
|
- sft |
|
|
base_model: mistralai/Mistral-7B-Instruct-v0.2 |
|
|
datasets: |
|
|
- generator |
|
|
model-index: |
|
|
- name: data |
|
|
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. --> |
|
|
|
|
|
# data |
|
|
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 1.1832 |
|
|
|
|
|
## 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: 0.0002 |
|
|
- train_batch_size: 1 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 4 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 1 |
|
|
- training_steps: 1000 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.9391 | 0.1479 | 25 | 0.6653 | |
|
|
| 0.6138 | 0.2959 | 50 | 0.6126 | |
|
|
| 0.6039 | 0.4438 | 75 | 0.6061 | |
|
|
| 0.5927 | 0.5917 | 100 | 0.5998 | |
|
|
| 0.5973 | 0.7396 | 125 | 0.5946 | |
|
|
| 0.602 | 0.8876 | 150 | 0.5943 | |
|
|
| 0.547 | 1.0355 | 175 | 0.6319 | |
|
|
| 0.4239 | 1.1834 | 200 | 0.6169 | |
|
|
| 0.4301 | 1.3314 | 225 | 0.6158 | |
|
|
| 0.4176 | 1.4793 | 250 | 0.6193 | |
|
|
| 0.4295 | 1.6272 | 275 | 0.6242 | |
|
|
| 0.4252 | 1.7751 | 300 | 0.6265 | |
|
|
| 0.4252 | 1.9231 | 325 | 0.6264 | |
|
|
| 0.3591 | 2.0710 | 350 | 0.6893 | |
|
|
| 0.2758 | 2.2189 | 375 | 0.7153 | |
|
|
| 0.2702 | 2.3669 | 400 | 0.7170 | |
|
|
| 0.2797 | 2.5148 | 425 | 0.7173 | |
|
|
| 0.2727 | 2.6627 | 450 | 0.7144 | |
|
|
| 0.2817 | 2.8107 | 475 | 0.7169 | |
|
|
| 0.2798 | 2.9586 | 500 | 0.7016 | |
|
|
| 0.1922 | 3.1065 | 525 | 0.8090 | |
|
|
| 0.16 | 3.2544 | 550 | 0.8373 | |
|
|
| 0.1623 | 3.4024 | 575 | 0.8372 | |
|
|
| 0.1632 | 3.5503 | 600 | 0.8402 | |
|
|
| 0.1618 | 3.6982 | 625 | 0.8558 | |
|
|
| 0.1732 | 3.8462 | 650 | 0.8581 | |
|
|
| 0.1687 | 3.9941 | 675 | 0.8611 | |
|
|
| 0.0961 | 4.1420 | 700 | 0.9902 | |
|
|
| 0.0879 | 4.2899 | 725 | 1.0102 | |
|
|
| 0.0899 | 4.4379 | 750 | 1.0345 | |
|
|
| 0.0899 | 4.5858 | 775 | 1.0256 | |
|
|
| 0.0882 | 4.7337 | 800 | 1.0273 | |
|
|
| 0.0893 | 4.8817 | 825 | 1.0559 | |
|
|
| 0.0824 | 5.0296 | 850 | 1.0753 | |
|
|
| 0.052 | 5.1775 | 875 | 1.1582 | |
|
|
| 0.052 | 5.3254 | 900 | 1.1643 | |
|
|
| 0.0526 | 5.4734 | 925 | 1.1923 | |
|
|
| 0.0497 | 5.6213 | 950 | 1.1759 | |
|
|
| 0.0496 | 5.7692 | 975 | 1.1812 | |
|
|
| 0.0477 | 5.9172 | 1000 | 1.1832 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.10.0 |
|
|
- Transformers 4.40.0 |
|
|
- Pytorch 2.2.2 |
|
|
- Datasets 2.19.0 |
|
|
- Tokenizers 0.19.1 |
