zhongshupeng/dataset_4090_2

# Disclaimer:
this dataset is curated for NeurIPS 2023 LLM efficiency challange, and currently work in progress. Please use at your own risk.

# Data composition:
All data were derived from the training set portion of the open source dataset.

**gsm2k_dolly15k_cnnadd6k_mmlulog1.7w_bbqabc8k.json**：

-gsm8k_2000: https://huggingface.co/datasets/gsm8k

-dolly_15000: https://huggingface.co/datasets/databricks/databricks-dolly-15k

-cnn_dailymail_6000: https://huggingface.co/datasets/cnn_dailymail

-mmlu_17000: https://huggingface.co/datasets/cais/mmlu

-bbq_8000: https://huggingface.co/datasets/tasksource/bigbench

**lima_4kall.json**

-lima_1000: https://huggingface.co/datasets/GAIR/lima

-3000 of gsm8k_dolly15k_cnnadd8k_mmlulog1.7w_bbqabc8k.json: https://huggingface.co/datasets/zhongshupeng/dataset_4090_1

原始信息汇总

数据集概述

数据组成

所有数据均来源于开源数据集的训练集部分。

数据文件

• gsm2k_dolly15k_cnnadd6k_mmlulog1.7w_bbqabc8k.json

  • gsm8k_2000
  • dolly_15000
  • cnn_dailymail_6000
  • mmlu_17000
  • bbq_8000

• lima_4kall.json

  • lima_1000
  • 3000 条数据来自 gsm8k_dolly15k_cnnadd8k_mmlulog1.7w_bbqabc8k.json