alibaba / alibaba/ChatLearn

[Alignment training (DPO)] Megatron support for QWEN2.5 series models

Open
#183 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
450
Forks
40
PR merge metrics
No merged PRs in 30d

Description

when will the QWEN2.5 series models support alignment training (RLHF、DPO、OnlineDPO、GRPO) using the Megatron

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by reviewing ChatLearn's existing alignment-training support and the entry points for Megatron and QWEN model integration. Determine the scope required for QWEN2.5 support across RLHF, DPO, OnlineDPO, and GRPO; done means these workflows operate with the QWEN2.5 series through Megatron.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.