5 Pull requests merged by 4 people

• ✒️ Fix typo in formatting_func's documentation in ConstantLengthDataset

  #2549 merged Jan 7, 2025

• 🧑‍🤝‍🧑 Proper metrics gathering across ranks before logging

  #2474 merged Jan 7, 2025

• ©️ Update copyrights year

  #2547 merged Jan 7, 2025

• 🚜 Use field in dataclasses

  #2494 merged Jan 6, 2025

• Remove graph breaks for torch.compile() in padding free branch in DataCollatorForCompletionOnlyLM

  #2158 merged Jan 6, 2025

2 Pull requests opened by 2 people

• MPO

  #2544 opened Jan 6, 2025

• [Judges] rlhflow pairwise judges

  #2548 opened Jan 7, 2025

1 Issue closed by 1 person

• DPOTrainer log metrics are not gathered and meaned across ranks

  #2468 closed Jan 7, 2025

2 Issues opened by 2 people

• Finetuning on the last turn of multi-turn conversations

  #2545 opened Jan 6, 2025

• Dataset type conversion utilities

  #2543 opened Jan 6, 2025

10 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• add xpu support for DPO

  #2533 commented on Jan 8, 2025 • 6 new comments

• 🕊️ DPO padding free

  #2520 commented on Jan 7, 2025 • 5 new comments

• add "_prepare_fsdp" for DPOTrainer

  #2539 commented on Jan 8, 2025 • 2 new comments

• plz make GPOTrainer! (Generalized Preference Optimization)

  #2028 commented on Jan 7, 2025 • 0 new comments

• [GRPO] initial GRPO trainer

  #1954 commented on Jan 5, 2025 • 0 new comments

• Add VAS to TRL

  #2195 commented on Jan 7, 2025 • 0 new comments

• [Liger] Integrate Liger CPO & SimPO

  #2506 commented on Jan 7, 2025 • 0 new comments

• RLOO trainer: fix calculations of steps, episodes and epochs

  #2516 commented on Jan 7, 2025 • 0 new comments

• [ORPO] revert orpo changes

  #2527 commented on Jan 6, 2025 • 0 new comments

• Issues Auto-Labeller

  #2542 commented on Jan 7, 2025 • 0 new comments