OFA-Sys/gsm8k-ScRel

Issues

• 关于MuggleMath的数据增强代码

  #24 opened 3 months ago by hlcle
  0

• About the collect_rejection_sampling.py

  #23 opened 7 months ago by jzh9830
  1

• if you have some plan to release data?

  #18 opened a year ago by ngc7292
  8

• Is MuggleMath dataset suitable for pre-training?

  #20 opened a year ago by JingyiWang3
  1

• The RFT data

  #19 opened a year ago by ZIKEYUAN
  3

• 70B training fails

  #16 opened a year ago by kumar-shridhar
  6

• Is majority voting(self-consistency) helpful for 70B llama2-sft model?

  #15 opened a year ago by platoonpluto
  1

• the inference of OFA-Sys/gsm8k-rft-llama13b2-u13b has shape error: 13Bllama2的u13b版本推理时出现矩阵形状错误

  #14 opened a year ago by AegeanYan
  8

• 加载作者开源的 OFA-Sys/gsm8k-rft-llama7b-u13b 报错

  #8 opened a year ago by Haskely
  12

• Enviroment

  #13 opened a year ago by nuochenpku
  3

• When will release 33b RFT model?

  #11 opened a year ago by nuochenpku
  4

• Reproducing llama7b2-sft problem

  #12 opened a year ago by huijiawu0
  2

• Release the RFT 7B model

  #2 opened a year ago by wenhuchen
  14

• When will release model of LLama13b RFT model?

  #10 opened a year ago by xingweiqu
  3

• problems about reproducing llama7b2-sft and llama7b2-rft-100

  #9 opened a year ago by ziyuwan
  23

• 关于源码的一些细节问题

  #7 opened a year ago by Haskely
  1

• test.py 里面use_diverse_beam 和do_sample 都默认是False?

  #5 opened a year ago by bestpredicts
  1

• Questions about RFT Inference

  #6 opened a year ago by waterhorse1
  5

• SFT

  #1 opened a year ago by eyuansu62
  3

• Release RFT datasets

  #4 opened a year ago by nuochenpku
  4

• Missing test.py file

  #3 opened a year ago by huijiawu0
  2