|
|
--- |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- tatsu-lab/alpaca |
|
|
language: |
|
|
- zh |
|
|
- en |
|
|
library_name: transformers |
|
|
tags: |
|
|
- baichuan |
|
|
--- |
|
|
|
|
|
An instruction-tuned LoRA model of https://huggingface.co/baichuan-inc/baichuan-7B |
|
|
|
|
|
This checkpoint is trained with: https://github.com/hiyouga/LLaMA-Efficient-Tuning |
|
|
|
|
|
Usage: |
|
|
```python |
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer |
|
|
from peft import PeftModel |
|
|
|
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained("baichuan-inc/baichuan-7B", trust_remote_code=True) |
|
|
model = AutoModelForCausalLM.from_pretrained("baichuan-inc/baichuan-7B", device_map="auto", trust_remote_code=True) |
|
|
model = PeftModel.from_pretrained(model, "chenliang1225/baichuan-7b-sft") |
|
|
streamer = TextStreamer(tokenizer, timeout=60.0, skip_prompt=True, skip_special_tokens=True) |
|
|
|
|
|
query = "陨石为什么总能落在陨石坑里？" |
|
|
|
|
|
inputs = tokenizer(["### Instruction:\n{}\n\n### Response:\n".format(query)], return_tensors="pt") |
|
|
inputs = inputs.to("cuda") |
|
|
generate_ids = model.generate(**inputs, max_new_tokens=256, streamer=streamer, top_p=0.7, temperature=0.95) |
|
|
``` |
|
|
|
|
|
|
