|
|
--- |
|
|
base_model: unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit |
|
|
tags: |
|
|
- text-generation-inference |
|
|
- transformers |
|
|
- unsloth |
|
|
- llama |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- fr |
|
|
- en |
|
|
datasets: |
|
|
- jpacifico/French-Alpaca-dataset-Instruct-55K |
|
|
--- |
|
|
|
|
|
# Uploaded finetuned model |
|
|
|
|
|
- **Developed by:** mintujohnson |
|
|
- **License:** apache-2.0 |
|
|
- **Finetuned from model :** unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit |
|
|
|
|
|
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. |
|
|
|
|
|
# Inference |
|
|
|
|
|
```python |
|
|
from unsloth import FastLanguageModel |
|
|
from transformers import TextStreamer |
|
|
|
|
|
|
|
|
model_path = "mintujohnson/Llama-3.2-3B-French-Instruct" |
|
|
model, tokenizer = FastLanguageModel.from_pretrained(model_name = model_path, max_seq_length = 128, |
|
|
dtype = None, load_in_4bit = True) |
|
|
|
|
|
def inference(messages, model, tokenizer): |
|
|
|
|
|
FastLanguageModel.for_inference(model) # Enable native 2x faster inference |
|
|
|
|
|
|
|
|
inputs = tokenizer.apply_chat_template( |
|
|
messages, tokenize = True, |
|
|
add_generation_prompt = True, # Must add for generation |
|
|
return_tensors = "pt", |
|
|
).to("cuda") |
|
|
print(tokenizer.decode(inputs[0], skip_special_tokens=False)) |
|
|
text_streamer = TextStreamer(tokenizer, skip_prompt = True) |
|
|
_ = model.generate( |
|
|
input_ids = inputs, streamer = text_streamer, max_new_tokens = 128, |
|
|
use_cache = True, temperature = 1.5, min_p = 0.1) |
|
|
|
|
|
messages = [ |
|
|
{"role": "user", "content": "où est la Normandie?"}, |
|
|
] |
|
|
output = inference(messages, model, tokenizer) |
|
|
``` |
|
|
|
|
|
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth) |
