请问数据集的来源是什么，请问可以提供微调过的mistral-7b-sft、math-shepherd-mistral-7b-prm模型吗 - Githubissues

openreasoner / openr

OpenR: An Open Source Framework for Advanced Reasoning with Large Language Models

https://openreasoner.github.io/

MIT License

1.04k stars 76 forks source link

请问数据集的来源是什么，请问可以提供微调过的mistral-7b-sft、math-shepherd-mistral-7b-prm模型吗 #36

Open Brainth opened 2 weeks ago

Brainth commented 2 weeks ago

System Info

(https://github.com/openreasoner/openr/blob/main/envs/MATH/dataset/test500.jsonl)请问这个数据集的来源是什么，是基准的数据集还是自己构造的

Who can help?

(https://github.com/openreasoner/openr/blob/main/envs/MATH/dataset/test500.jsonl)请问这个数据集的来源是什么，是基准的数据集还是自己构造的

Information

[ ] The official example scripts
[ ] My own modified scripts

Tasks

[ ] An officially supported task in the codebase (such as scrips/, ...)
[ ] My own task or dataset (give details below)

Reproduction

(https://github.com/openreasoner/openr/blob/main/envs/MATH/dataset/test500.jsonl)请问这个数据集的来源是什么，是基准的数据集还是自己构造的

Expected behavior

(https://github.com/openreasoner/openr/blob/main/envs/MATH/dataset/test500.jsonl)请问这个数据集的来源是什么，是基准的数据集还是自己构造的

ziyuwan commented 1 week ago

Hi @Brainth The math-500 test set originated from OpenAI's paper Let's verify step by step. While mistral-7b-sft and math-shepherd-mistral-7b-prm are are from the paper of Math Shepherd, we didn't fine-tune it.