llm-jp-3-13b-it_lora-DPO-12-16

Uploaded model

• Developed by: thesugar
• License: CC-BY-NC-SA
• Finetuned from model : llm-jp/llm-jp-3-13b

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

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

HOW TO INFERENCE for competition evaluators

Google Colab L4 で実行

!pip install unsloth
!pip uninstall unsloth -y && pip install --upgrade --no-cache-dir --no-deps git+https://github.com/unslothai/unsloth.git

HF_TOKEN = # WRITE YOUR HF_TOKEN
ELYZA_TASKS_100_TV_JSONL_PATH = # WRITE
# Output for elyza-tasks-100-tv is saved as "output.jsonl"


from huggingface_hub import login
login(HF_TOKEN)

from unsloth import FastLanguageModel
import torch

max_seq_length = 2048
dtype = torch.bfloat16
load_in_4bit = True
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "thesugar/llm-jp-3-13b-it_lora-DPO-12-16",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
    token = HF_TOKEN,
)

import json
datasets = []
with open(ELYZA_TASKS_100_TV_JSONL_PATH, "r") as f:
    item = ""
    for line in f:
      line = line.strip()
      item += line
      if item.endswith("}"):
        datasets.append(json.loads(item))
        item = ""

from tqdm import tqdm

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 = 2048, 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})

with open("output.jsonl", "w") as f:
  for r in results:
    f.write(json.dumps(r, ensure_ascii=False) + "\n")

Development steps

• llm-jp/llm-jp-3-13b を量子化
• インストラクションチューニング
  • ichikara-instruction データセットの ichikara-instruction-003-001-1.json の全データを使用
• direct policy optimization
  • weblab-GENIAC/aya-ja-nemotron-dpo-masked からランダムに選択した 100 レコードを使用

Used datasets　and their licenses

ichikara-instruction: LLMのための日本語インストラクションデータ

• LLMのための日本語インストラクションデータ 公開ページ – LIAT-AIP homepage

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) CC-BY-NC-SA

weblab-GENIAC/aya-ja-nemotron-dpo-masked

• creator
• repository

weblab-GENIAC
weblab-GENIAC/aya-ja-nemotron-dpo-masked
Apache License 2.0