Issues

• Different dtype of model and ref_model

  RLHFlow/Online-DPO-R1#16 · Filter by author pspdada opened on Apr 14, 2025

• Code update request

  RLHFlow/Online-DPO-R1#15 · Filter by author Jiangchenglin521 opened on Apr 11, 2025
  1

• What is the correct prompt type for deepseek-ai/DeepSeek-R1-Distill-Llama-8B?

  RLHFlow/Online-DPO-R1#13 · Filter by author CurryxIaoHu opened on Apr 1, 2025
  1

• Garbled Outputs from vLLM Sampling Causing Instability in Preference Learning

  RLHFlow/Online-DPO-R1#12 · Filter by author pspdada opened on Mar 31, 2025
  1

• Questions about Dataset and Experiments

  RLHFlow/Online-DPO-R1#8 · Filter by author LLMforScience opened on Mar 9, 2025
  8

• <string>:1: Syntax warning: 'int' object is not callable; perhaps you missed a comma?

  RLHFlow/Online-DPO-R1#5 · Filter by author CurryxIaoHu opened on Feb 20, 2025
  2

• Did you do any performance comparison between online DPO and GRPO?

  RLHFlow/Online-DPO-R1#1 · Filter by author zhengyunqq opened on Feb 17, 2025
  1