README.md

---
base_model: llm-jp/llm-jp-3-13b
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
language:
- ja
license: cc-by-nc-sa-4.0
---
# はじめに
これは，東京大学松尾・岩澤研究室のLLM講座2024のコンペティションで提出するためのモデルです．

[llm-jp/llm-jp-3-13b](https://huggingface.co./llm-jp/llm-jp-3-13b)に，QLoRAによるSFTを施して，LoRAアダプタのみをこちらにアップしています．

chat templateは，[weblab-GENIAC/Tanuki-8B-dpo-v1.0](https://huggingface.co./weblab-GENIAC/Tanuki-8B-dpo-v1.0)のものと同一のものを使用しています．

# 推論方法
提供された環境で，以下のように推論します．L4 GPU×1のインスタンスで，vLLMを用いて推論します．

Jupyter Notebookで，一かたまりごとに一つのセルになっています．順番に実行してください．
```python
!pip uninstall numpy -y
!pip install numpy==1.26.4

%%time
%pip install vllm==0.6.4.post1 --force-reinstall

!pip install ipywidgets

import time
import torch
import transformers
from transformers import (
    AutoTokenizer,
    AutoModelForCausalLM,
    BitsAndBytesConfig
)
import vllm 
from vllm.lora.request import LoRARequest
from jinja2 import Template
print(vllm.__version__)

MAX_LENGTH = 1024
MODEL_NAME = "llm-jp/llm-jp-3-13b"
print(MODEL_NAME)

import os
os.environ["HF_TOKEN"] = "あなたのHugging Faceトークン"

from vllm.lora.request import LoRARequest
llm = vllm.LLM(
    MODEL_NAME, 
    tensor_parallel_size=1, 
    gpu_memory_utilization=0.95,
    trust_remote_code=True,
    enforce_eager=True,
    max_model_len=MAX_LENGTH,
    enable_lora=True,
    quantization="bitsandbytes",
    load_format="bitsandbytes"
)
tokenizer = llm.get_tokenizer()

from transformers import AutoTokenizer
sft_tokenizer = AutoTokenizer.from_pretrained(
    "weblab-GENIAC/Tanuki-8B-dpo-v1.0"
)
tokenizer.chat_template = sft_tokenizer.chat_template

from huggingface_hub import snapshot_download
lora_path = snapshot_download(repo_id="OsakanaTeishoku/1204lora")

from datasets import load_dataset
data_files = {"test": "elyza-tasks-100-TV_0.jsonl"}  
tasks = load_dataset("json", data_files=data_files, split="test")

messages_list = [
    [{"role": "user", "content": tasks["input"][i]}] for i in range(len(tasks))
]
prompts = [line[0]["content"] for line in messages_list]
prompt_token_ids = [tokenizer.apply_chat_template(messages, add_generation_prompt=True) for messages in messages_list]
sampling_params = vllm.SamplingParams(
    temperature=0.7,
    max_tokens=1024,
    repetition_penalty=1.05,
    top_p=0.9,
)
outputs = llm.generate(
    prompt_token_ids=prompt_token_ids,
    sampling_params=sampling_params,
    lora_request=LoRARequest("lora", 1, lora_path), # LoRA adapter
)
for prompt, response in zip(prompts, outputs):
    print("prompt:", prompt)
    print("output:", response.outputs[0].text.strip())
    print("-"*80)   
import json
data = [{
    "task_id": i, 
    #"input": prompts[i],
    "output": outputs[i].outputs[0].text.strip()
} for i in range(len(tasks))]
file_path_with_unicode = 'output.jsonl'
with open(file_path_with_unicode, 'w', encoding='utf-8') as file:
    for entry in data:
        json.dump(entry, file, ensure_ascii=False)
        file.write('\n')
print(f"Saved json {file_path_with_unicode} !")
```
# Change log
- 2024/12/26: 推論コードの余分なコメント部分の削除，リンクの追加

# Uploaded  model

- **Developed by:** OsakanaTeishoku
- **License:** cc-by-nc-sa-4.0
- **Finetuned from model :** llm-jp/llm-jp-3-13b

This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)