monamonamona/llm-jp-3-13b-finetune-241124

Uploaded model

• Developed by: monamonamona
• License: apache-2.0
• Finetuned from model : llm-jp/llm-jp-3-13b

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.

Sample Use

以下はelyza-tasks-100-TV_0.jsonlの回答のためのコードです。

# 学習したモデルを用いてタスクを実行
from tqdm import tqdm

# 推論するためにモデルのモードを変更
FastLanguageModel.for_inference(model)

results = []
for dt in tqdm(datasets):
    
    input = dt["input"]
    
    prompt = f"""### 指示\n{input}\n### 回答\n"""
    
    inputs = tokenizer([prompt], return_tensors = "pt").to(model.device)
    outputs = model.generate(**inputs, max_new_tokens = 512, use_cache = True, do_sample=False, repetition_penalty=1.2)
    
    prediction = tokenizer.decode(outputs[0], skip_special_tokens=True).split('\n### 回答')[-1]
    
    results.append({"task_id": dt["task_id"], "input": input, "output": prediction})