kamalkraj/e5-mistral-7b-instruct

Issues

• Is gradient accumulation applicable for contrastive learning?

  #15 opened 4 months ago by Hypothesis-Z
  0

• Exception when "--checkpointing_steps" is set

  #12 opened 9 months ago by Hypothesis-Z
  2

• OVERFLOW! Rank 0 Skipping step

  #14 opened 7 months ago by liujiqiang999
  0

• Best Practices for Fine-Tuning Models on Multi-Hop Datasets?

  #13 opened 7 months ago by Leon-Sander
  0

• Am I using the code incorrectly? help me

  #6 opened a year ago by Pang-dachu
  14

• ValueError: expected sequence of length 595 at dim 1 (got 589)

  #11 opened 9 months ago by Hypothesis-Z
  1

• OOM with 2 GPUs (48GB in total)

  #10 opened 10 months ago by yurinoviello
  0

• when I modify batch_size larger than 1

  #3 opened a year ago by sangzisen
  8

• Inference OOM with 80G GPU 1000 input sequence length

  #7 opened a year ago by charliedream1
  6

• Is there any way to use multiple GPUs?

  #5 opened a year ago by Pang-dachu
  1

• How much GPU should it need minimum to run this?

  #4 opened a year ago by anonymousz97
  2

• out of memory on A100-80G.

  #2 opened a year ago by Joris-Fu
  7

• Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu!

  #1 opened a year ago by Tostino
  9