Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

在llama-2-13b上使用huangyt/FINETUNE2_TEST資料集進行訓練,總資料筆數約2.2w

Fine-Tuning Information

Fine-Tuning Detail

Evaluation

Model Average ARC HellaSwag MMLU TruthfulQA
meta-llama/Llama-2-13b-hf 56.9 58.11 80.97 54.34 34.17
meta-llama/Llama-2-13b-chat-hf 59.93 59.04 81.94 54.64 44.12
CHIH-HUNG/llama-2-13b-FINETUNE2_TEST_2.2w 58.46 56.23 82.7 55.35 39.55

How to convert dataset to json

import json
from datasets import load_dataset

# 讀取數據集,take可以取得該數據集前n筆資料
dataset = load_dataset("huangyt/FINETUNE2_TEST", split="train", streaming=True)

# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
    extracted_example = {
        "instruction": example["instruction"],
        "input": example["input"],
        "output": example["output"]
    }
    extracted_data.append(extracted_example)

# 指定 JSON 文件名稱
json_filename = "FINETUNE2_TEST.json"

# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
    json.dump(extracted_data, json_file, indent=4)

print(f"數據已提取並保存為 {json_filename}")