README.md

Llama3.2-3B LoRA Fine-tuned Model

This is a LoRA fine-tuned version of Llama3.2-3B on https://huggingface.co/datasets/liumindmind/NekoQA-10K.

How to Use

messages = [
    {"role" : "user", "content" : "摸摸"}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize = False,
    add_generation_prompt = True,
)

from transformers import TextStreamer
_ = model.generate(
    **tokenizer(text, return_tensors = "pt").to("cuda"),
    max_new_tokens = 1500,
    temperature = 1.0, top_p = 0.8, top_k = 20,
    streamer = TextStreamer(tokenizer, skip_prompt = True),
)