baaivision/JudgeLM

Issues

• KeyError: 'scores' in judgelm_preprocess.py

  #22 opened a month ago by ilyes-c
  0

• Can you share theminimal `.jsonl` file examples

  #17 opened 6 months ago by deshwalmahesh
  1

• Is this project still under active deveolopment?

  #21 opened 6 months ago by nkwangleiGIT
  1

• Issue running single judgement with references

  #20 opened 6 months ago by sersoage
  0

• How to prompt for single answer without reference?

  #16 opened 7 months ago by andrewrreed
  1

• 训练结果复现不符合预期

  #19 opened 7 months ago by HuihuiChyan
  0

• Leaderboard of JudgLM evaluations

  #15 opened 7 months ago by sachith-surge
  2

• How to make JudgeLM Generate Chinese Output?

  #18 opened 7 months ago by realbillbao
  0

• 关于中文问答的自动评测

  #14 opened 7 months ago by Lam1360
  2

• 您好 你们在训练33b的时候用了多少资源呀

  #13 opened 7 months ago by eyuansu62
  2

• HuggingFace TGI for fast inference and serving

  #12 opened 8 months ago by ogencoglu
  2

• Have you done any experiments to prove the performance of the model under ground truth labeling and not GPT4 labeling

  #11 opened 8 months ago by ZiruiSongBest
  2

• 你好，请教下如何同时给多个答案打分呢？看目前的代码好像还不支持？

  #5 opened 8 months ago by TUNATAKA
  3

• preprocess

  #3 opened 8 months ago by darinkishore
  6

• hi, how can we use the judge_pair template to judge a single answer.

  #4 opened 8 months ago by eyuansu62
  5

• Can you describe a bit more about the training process and data strategy. Load given to Inference in the paper

  #2 opened 8 months ago by deshwalmahesh
  2