Xirider/finetune-gpt2xl

Issues

• TypeError: __init__() got an unexpected keyword argument 'no_args_is_help'

  #24 opened 2 years ago by SeekPoint
  0

• AttributeError: 'DeepSpeedCPUAdam' object has no attribute 'ds_opt_adam'

  #16 opened 3 years ago by remotejob
  8

• fine tuning GPT-J 6B?

  #22 opened 2 years ago by silvacarl2
  1

• IndexError: index out of bounds

  #23 opened 2 years ago by GreenTeaBD
  1

• Using [text,labels] instead of just [text] in Datasets

  #21 opened 2 years ago by imthebilliejoe
  1

• subprocess.CalledProcessError:

  #20 opened 3 years ago by Dhanachandra
  1

• Out of memory with RTX3090

  #19 opened 3 years ago by PyxAI
  4

• Suspected optimizer issue causing crashes

  #3 opened 4 years ago by kinoc
  1

• Feeding the model separate examples instead of one continuous block of text

  #17 opened 3 years ago by CupOfGeo
  1

• Gpt-neo inference with Deepspeed: IndexError: Dimension out of range

  #15 opened 3 years ago by kingpalethe
  3

• New issue with Pandas

  #14 opened 3 years ago by barakw2021
  0

• Crashes with new Transformers version

  #13 opened 4 years ago by barakw2021
  1

• Can't change BOS token or EOS token for GPT Neo

  #12 opened 4 years ago by mallorbc
  3

• Ideal number of epochs? Number of examples meaning?

  #11 opened 4 years ago by mallorbc
  1

• Freezing at "Using /home/user/.cache/torch_extensions as PyTorch extensions root..."

  #10 opened 4 years ago by mallorbc
  4

• TypeError: unsupported operand type(s) for -: 'float' and 'str' on AWS g4dn.12xlarge

  #8 opened 4 years ago by sibeshkar
  0

• Resume from checkpoint

  #9 opened 4 years ago by ArturTanona
  2

• Errors while trying to train with two GPUs

  #7 opened 4 years ago by barakw2021
  4

• Unable to proceed, no GPU resources available

  #6 opened 4 years ago by bpm246
  3

• Multiple entries csv

  #5 opened 4 years ago by kikirizki
  1

• Training on a larger dataset fails due to memory issues on faster GPUs

  #4 opened 4 years ago by jonnyplatt
  3

• Exception: Installed CUDA version 11.0 does not match the version torch was compiled with 11.1 [SOLUTION]

  #1 opened 4 years ago by CupOfGeo
  2