Issues

• RRHF with Online Sampling

  GanjinZero/RRHF#57 · Filter by author sqqiao opened on Sep 14, 2024
  1

• resize embedding after add_special_tokens

  GanjinZero/RRHF#56 · Filter by author Switchsyj opened on Sep 3, 2024

• Runtime error：数据类型报错

  GanjinZero/RRHF#55 · Filter by author sqqiao opened on Jul 23, 2024

• 关于ppl的方差

  GanjinZero/RRHF#54 · Filter by author skepsun opened on Feb 1, 2024
  1

• 算loss的时候求均值的时候是不是可以优化

  GanjinZero/RRHF#51 · Filter by author shyoulala opened on Dec 21, 2023
  6

• 数据构造问题

  GanjinZero/RRHF#50 · Filter by author lylcst opened on Dec 11, 2023
  1

• bug 计算sft损失的时候

  GanjinZero/RRHF#48 · Filter by author shyoulala opened on Nov 24, 2023
  2

• 关于alpaca-7B和LLaMA-7B

  GanjinZero/RRHF#47 · Filter by author NEUBuffett opened on Oct 22, 2023
  3

• 有关IMDB数据集的问题

  GanjinZero/RRHF#46 · Filter by author stevie1023 opened on Oct 12, 2023
  2

• 加载模型的问题

  GanjinZero/RRHF#43 · Filter by author LiangZhuuu opened on Sep 6, 2023
  11

• 训练过程OOM的问题

  GanjinZero/RRHF#41 · Filter by author Guochry opened on Aug 21, 2023
  1

• Wombat与RRHF

  GanjinZero/RRHF#40 · Filter by author Guochry opened on Aug 2, 2023
  4

• labels != -100的作用是什么

  GanjinZero/RRHF#38 · Filter by author LSX-Sneakerprogrammer opened on Jul 19, 2023
  3

• RRHFTrainer.gather_logits_labels label in-place operation error

  GanjinZero/RRHF#37 · Filter by author asadfgglie opened on Jul 19, 2023
  8

• 评估方法与位置有很大关系

  GanjinZero/RRHF#32 · Filter by author xiaoyuan1996 opened on Jun 6, 2023
  2

• 期待LoRA或ptuning

  GanjinZero/RRHF#31 · Filter by author Noyce765103 opened on May 31, 2023
  1

• How to use it. Is there some code examples?

  GanjinZero/RRHF#28 · Filter by author Spider4U opened on May 16, 2023
  1

• 对于重复score答案样本的处理疑问

  GanjinZero/RRHF#25 · Filter by author yanhan19940405 opened on May 8, 2023
  7

• loss的代码关于batch size的处理有bug。

  GanjinZero/RRHF#23 · Filter by author echoht opened on May 5, 2023
  4

• training with my own gpt2

  GanjinZero/RRHF#22 · Filter by author dyyzhmm opened on May 3, 2023
  1

• PPO implementation

  GanjinZero/RRHF#19 · Filter by author yuzc19 opened on Apr 25, 2023
  2

• Wombat-7B，Wombat-7B-gpt4 and ChatGPT Results on Comparison based on Vicuna test set, evaluation by gpt-4.

  GanjinZero/RRHF#18 · Filter by author onlyfish79 opened on Apr 24, 2023
  4

• Results on Comparison based on Vicuna test set

  GanjinZero/RRHF#16 · Filter by author LeeShiyang opened on Apr 19, 2023
  1

• Why use HingeLoss instead of BPRLoss ?

  GanjinZero/RRHF#15 · Filter by author KID-22 opened on Apr 19, 2023
  1

• This loss seems to consume a lot of memory.

  GanjinZero/RRHF#13 · Filter by author piekey1994 opened on Apr 14, 2023
  4