Neito8/llm-jp-3-13b-it_lora

Uploaded model

• Developed by: Neito8
• 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.

Usage

Execute following code in Google Colab

# 必要なライブラリをインストール
!pip install unsloth
!pip uninstall unsloth -y && pip install --upgrade --no-cache-dir "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
!pip install -U torch
!pip install -U peft
# 必要なライブラリを読み込み
from unsloth import FastLanguageModel
from peft import PeftModel
import torch
import json
from tqdm import tqdm
import re
# ベースとなるモデル。
model_id = "llm-jp/llm-jp-3-13b"
from google.colab import userdata
HF_TOKEN=userdata.get('HF_TOKEN')
# unslothのFastLanguageModelで元のモデルをロード。
dtype = None # Noneにしておけば自動で設定
load_in_4bit = True # 今回は13Bモデルを扱うためTrue
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name=model_id,
    dtype=dtype,
    load_in_4bit=load_in_4bit,
    trust_remote_code=True,
)
# SFT用のモデルを用意。
model = FastLanguageModel.get_peft_model(
    model,
    r = 32,
    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
                      "gate_proj", "up_proj", "down_proj",],
    lora_alpha = 32,
    lora_dropout = 0.05,
    bias = "none",
    use_gradient_checkpointing = "unsloth",
    random_state = 3407,
    use_rslora = False,
    loftq_config = None,
    max_seq_length = max_seq_length,
)
# タスクとなるデータの読み込み。
# 事前にデータをアップロードしてください。
datasets = []
with open("./elyza-tasks-100-TV_0.jsonl", "r") as f:
    item = ""
    for line in f:
      line = line.strip()
      item += line
      if item.endswith("}"):
        datasets.append(json.loads(item))
        item = ""
# モデルを用いてタスクの推論。
# 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} 簡潔に回答してください \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]
  prediction = re.sub(r"[*#]", "", prediction)
  results.append({"task_id": dt["task_id"], "input": input, "output": prediction})
# 結果をjsonlで保存。
json_file_id = re.sub(".*/", "", adapter_id)
with open(f"/content/{json_file_id}_output.jsonl", 'w', encoding='utf-8') as f:
    for result in results:
        json.dump(result, f, ensure_ascii=False)
        f.write('\n')

Datasets

Instruction tuning

The models have been fine-tuned on the following datasets.

Language	Dataset	description
Japanese	ichikara-instruction-003-001-1.json	A manually constructed instruction dataset
	Synthesized data from Elyza-tasks-100	Synthesize data from Elyza-tasks-100 by using LLM(Tanuki-8x8B)

データセット作成チーム： 関根聡, 安藤まや, 後藤美知子, 鈴木久美, 河原大輔, 井之上直也, 乾健太郎. ichikara-instruction: LLMのための日本語インストラクションデータの構築. 言語処理学会第30回年次大会(2024)