openai/weak-to-strong

Issues

• How to use GPT-4 as strong model

  #35 opened 10 months ago by sr5434
  1

• Why are only binary classification tasks tested for NLP?

  #37 opened 8 months ago by sev777
  0

• Training stopping criteria: fixed epochs ?

  #36 opened 8 months ago by Aakriti05
  0

• Some questions about the anthropic_hh dataset results

  #34 opened 10 months ago by rgreenblatt
  0

• Machine learning newbie hopes to get help

  #24 opened a year ago by snowkcon
  5

• I created a code for "autonomous AI". But since I'm an amateur, I don't know how to implement it.

  #30 opened a year ago by LineaLinelis
  0

• Unexpected keyword argument 'bf16'

  #11 opened a year ago by agokrani
  14

• UnboundLocalError: local variable 'minibatch_size' referenced before assignment

  #25 opened a year ago by philipkd
  0

• Exploring Weak to Strong Generalization from a pre-training standpoint

  #12 opened a year ago by rokosbasilisk
  4

• hi guys!

  #1 opened a year ago by MadcowD
  6

• Will you guys be releasing code for step 2 of the methodology "Train a strong student model with weak supervision."

  #19 opened a year ago by hoalarious
  2

• Preprocessed Chess Puzzle Data

  #10 opened a year ago by y12uc231
  4

• what's the meaning of fraction of GPT4?

  #17 opened a year ago by ljch2018
  2

• Observing eval accuracy considerably lower than reported?

  #3 opened a year ago by knagrecha
  12

• The schedule.step() should be called outside the dataloader loop.

  #15 opened a year ago by wangguanan
  3

• Protect main branch and make changes through pull requests

  #6 opened a year ago by EwoutH
  1

• Some thoughts

  #13 opened a year ago by SEU-zxj
  1