|
|
--- |
|
|
license: mit |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- mistralai/Mistral-7B-v0.1 |
|
|
pipeline_tag: question-answering |
|
|
tags: |
|
|
- technology |
|
|
- QA |
|
|
--- |
|
|
# TechChat |
|
|
|
|
|
**TechChat** is a domain-specific chatbot fine-tuned from [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) using LoRA. |
|
|
|
|
|
## Model Details |
|
|
- **Base model:** mistralai/Mistral-7B-v0.1 |
|
|
- **Fine-tuning method:** LoRA (Low-Rank Adaptation) |
|
|
- **Max sequence length:** 512 tokens |
|
|
|
|
|
## Intended Use |
|
|
- Technical Q&A in [your domain] |
|
|
- Chat-style interactions |
|
|
|
|
|
## Example |
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
|
|
model = AutoModelForCausalLM.from_pretrained("hari7261/TechChat") |
|
|
tokenizer = AutoTokenizer.from_pretrained("hari7261/TechChat") |
|
|
|
|
|
prompt = "Explain DNS in simple terms." |
|
|
inputs = tokenizer(prompt, return_tensors="pt") |
|
|
outputs = model.generate(**inputs, max_new_tokens=150) |
|
|
print(tokenizer.decode(outputs[0], skip_special_tokens=True)) |
