princeton-nlp/ProLong

Issues

• Data Recipe

  #2 opened 15 days ago by lwang2070
  1

• S3 Bucket Access

  #7 opened 15 days ago by gokulr-cerebras
  5

• Data Processing

  #8 opened 15 days ago by jialianwww
  1

• Fine-tuning 64k OOM

  #10 opened 15 days ago by xjwhy
  2

• How to pack prolong-ultrachat-64K data with another model's tokenizer?

  #11 opened 15 days ago by ZetangForward
  1

• Streaming Dataset

  #12 opened a month ago by JialianW
  3

• Multi-node Training

  #9 opened 4 months ago by chtmp223
  2

• What's the change to Llama model ?

  #6 opened 5 months ago by wtzhang99
  1

• Error while loading ultrachat dataset used in ProLong

  #5 opened 5 months ago by amy-hyunji
  2

• Specifying epochs instead of steps

  #4 opened 5 months ago by lilakk
  2

• [Bug] A tensor index bug in LLama Attention

  #3 opened 6 months ago by Chenfeng1271
  1

• Code Release

  #1 opened 6 months ago by wangxidong06
  0