Commits

Branch selector

main

User selector

All users

All time

Commit history

Commits on Mar 31, 2025

• [fix] Add param to resolve custom model loading failure (#845)

  BearBiscuit05

  authored
  1f78e8bView commit details

  Copy full SHA for 1f78e8b

  Browse repository at this point

• [doc] fix: typo for REINFORCE (#846)

  tongyx361

  authored
  a03a72aView commit details

  Copy full SHA for a03a72a

  Browse repository at this point

• [example] rollout: add vllm 0.8.2 mutli nodes generation bash (#838)

  BearBiscuit05

  authored
  7646e08View commit details

  Copy full SHA for 7646e08

  Browse repository at this point

• [BREAKING config] fix: move val_before_train to config yaml. Using trainer.val_before_train instead of +trainer.val_before_train going forward (#820)

  tongyx361

  authored
  64bddb6View commit details

  Copy full SHA for 64bddb6

  Browse repository at this point

Commits on Mar 30, 2025

• [BREAKING config] feat: add mlflow val generation log and uri config (#822)

  Show description for 7fbf609
  ccclyu

  authored
  7fbf609View commit details

  Copy full SHA for 7fbf609

  Browse repository at this point

• docs: add PURE to README.md (#826)

  Show description for 0e99caa
  CJReinforce

  authored
  0e99caaView commit details

  Copy full SHA for 0e99caa

  Browse repository at this point

• [sglang] fix: add memory saver support to sglang rollout to avoid OOMs (#756)

  Show description for 5138a22
  

  SwordFaith

  and

  ocss884

  authored
  5138a22View commit details

  Copy full SHA for 5138a22

  Browse repository at this point

• Megatron checkpoint default not save hf_models, and provide model merge tool. (#780)

  Show description for ccab836
  ETOgaosion

  authored
  ccab836View commit details

  Copy full SHA for ccab836

  Browse repository at this point

Commits on Mar 29, 2025

• Fix typo on installation guide (#813)

  Show description for 797f999
  dingyuan-shi

  authored
  797f999View commit details

  Copy full SHA for 797f999

  Browse repository at this point

• [misc] add deepseek v3 flops compute func (#814)

  BearBiscuit05

  authored
  0cf4ca4View commit details

  Copy full SHA for 0cf4ca4

  Browse repository at this point

• [megatron] fix: remove redundant return value for hf_config (#722)

  caaatch22

  authored
  f3913d0View commit details

  Copy full SHA for f3913d0

  Browse repository at this point

• docs: update checkpoint doc (#800)

  Show description for 50cba4a
  ETOgaosion

  authored
  50cba4aView commit details

  Copy full SHA for 50cba4a

  Browse repository at this point

Commits on Mar 28, 2025

• ci/cd: add pylint to CI (#811)

  Show description for 4f32b32
  frederrx

  authored
  4f32b32View commit details

  Copy full SHA for 4f32b32

  Browse repository at this point

• [trainer] fix: skip the update step when encountering gradient overflow (#789)

  Show description for 093e959
  zhao9797

  authored
  093e959View commit details

  Copy full SHA for 093e959

  Browse repository at this point

• Fix padding length for sglang rollout in veRL (#773)

  Show description for 52e80fc
  

  GeLee-Q

  and

  GeLee-Q

  authored
  52e80fcView commit details

  Copy full SHA for 52e80fc

  Browse repository at this point

• Update README.md (#797)

  Show description for 36a0f06
  PzySeere

  authored
  36a0f06View commit details

  Copy full SHA for 36a0f06

  Browse repository at this point

• Fix: Undefined variable name in the print_rank_0 statement (#788)

  Show description for 44ac51f
  jinqinn

  authored
  44ac51fView commit details

  Copy full SHA for 44ac51f

  Browse repository at this point

Commits on Mar 27, 2025

• refactor: remove redundant response_mask computation (#792)

  tongyx361

  authored
  f09e365View commit details

  Copy full SHA for f09e365

  Browse repository at this point

• doc: upgrade to vllm 0.8.2 (#769)

  hiyouga

  authored
  1b8b0e8View commit details

  Copy full SHA for 1b8b0e8

  Browse repository at this point

• In the GRPO example scripts, modify the `entropy_coeff` to 0 to ensure consistency with the GRPO paper. (#781)

  Show description for aab0176
  The-Hierophant

  authored
  aab0176View commit details

  Copy full SHA for aab0176

  Browse repository at this point

• prevent division by zero in SFT (#786)

  cpchenpi

  authored
  076acdfView commit details

  Copy full SHA for 076acdf

  Browse repository at this point

• fix: prevent NaN when all items are intentionally masked by adding epsilon to masked_mean (#782)

  The-Hierophant

  authored
  e9bcd2bView commit details

  Copy full SHA for e9bcd2b

  Browse repository at this point

• fix: set torch dtype to auto (#749)

  Show description for a1dd922
  ccclyu

  authored
  a1dd922View commit details

  Copy full SHA for a1dd922

  Browse repository at this point

Commits on Mar 26, 2025

• fix: prompt_token_ids should be list[int] instead of np.array (#772)

  Show description for fc27caf
  

  0x404

  and

  hiyouga

  authored
  fc27cafView commit details

  Copy full SHA for fc27caf

  Browse repository at this point

• Update math_dataset.py to fix typo in the annotation (#765)

  Show description for 591a476
  alexanderliu-creator

  authored
  591a476View commit details

  Copy full SHA for 591a476

  Browse repository at this point

• [Bug fix] change max_model_len to be configurable in vllm_rollout_spmd (#677)

  Show description for 7f10e57
  

  Tian Wang

  authored
  7f10e57View commit details

  Copy full SHA for 7f10e57

  Browse repository at this point

• [Feat] add max_ckpt_to_keep for old ckpt removal (#724)

  Show description for afb9f9f
  guanning03

  authored
  afb9f9fView commit details

  Copy full SHA for afb9f9f

  Browse repository at this point

• docs: doc improvements via Openhands, add SimpleRL-Zoo (#764)

  eric-haibin-lin

  authored
  9b5ffe4View commit details

  Copy full SHA for 9b5ffe4

  Browse repository at this point

• docs: support AMD (Rocm Kernel) - Merge upstream changes and update AMD tutorial (#741)

  Show description for 2970156
  

  yushengsu-thu

  and

  eric-haibin-lin

  authored
  2970156View commit details

  Copy full SHA for 2970156

  Browse repository at this point

Commits on Mar 25, 2025

• docker: Add dockerfile to build container for AWS Sagemaker training job (#763)

  BaiqingL

  authored
  8ee3faaView commit details

  Copy full SHA for 8ee3faa

  Browse repository at this point

• chore(deps): bump sglang[all] from 0.4.3.post3 to 0.4.4 (#646)

  Show description for deae725
  dependabot[bot]

  authored
  deae725View commit details

  Copy full SHA for deae725

  Browse repository at this point

• fix: prime refactor ignores extra_info (#717)

  ganler

  authored
  8594627View commit details

  Copy full SHA for 8594627

  Browse repository at this point

• docs: fix sglang installation rendering (#762)

  Show description for 6199b8c
  eric-haibin-lin

  authored
  6199b8cView commit details

  Copy full SHA for 6199b8c

  Browse repository at this point

• recipe: add reproducible PRIME baseline (#753)

  Show description for 0415b27
  ZefanW

  authored
  0415b27View commit details

  Copy full SHA for 0415b27

  Browse repository at this point

• [docs] Update the doc for vllm >= 0.8 (#755)

  Show description for fa18360
  BearBiscuit05

  authored
  fa18360View commit details

  Copy full SHA for fa18360

  Browse repository at this point