jongwooko/distillm

Issues

• Dataset link seems invalid now

  #12 opened 2 months ago by aaab8b
  2

• Problem during finetuning qwen2 models

  #13 opened 2 months ago by j-datta
  0

• Download the training/evaluation data

  #10 opened 4 months ago by ypw-lbj
  2

• gdk mismatch in equation?

  #9 opened 5 months ago by ogimgio
  0

• incorrect loss functions based on KLD definition

  #7 opened 5 months ago by mohammadreza-molapanah
  4

• the code of task-specific experiments

  #6 opened 6 months ago by wangfan120
  0

• Inconsistnet with your reported scores of teacher models

  #5 opened 8 months ago by liuxy1103
  12

• The metric of GPT-4 Eval

  #2 opened 8 months ago by liuxy1103
  1

• How to download the models after sft

  #3 opened 8 months ago by liuxy1103
  4

• OPTS+=" --kd-ratio 1.0"

  #4 opened 9 months ago by liuxy1103
  2