Commit Graph

3762 Commits (e0c68ab6d3d64f401208d6ec66815995cee233c3)
 

Author SHA1 Message Date
wangbluo 5ecc27e150 fix
2 months ago
wangbluo f98384aef6 fix
2 months ago
Hongxin Liu 646b3c5a90
[shardformer] fix linear 1d row and support uneven splits for fused qkv linear (#6084)
2 months ago
wangbluo b635dd0669 fix
2 months ago
wangbluo 3532f77b90 fix
2 months ago
wangbluo 3fab92166e fix
2 months ago
binmakeswell f4daf04270
add funding news (#6072)
2 months ago
wangbluo 6705dad41b fix
2 months ago
wangbluo 91ed32c256 fix
2 months ago
wangbluo 6fb1322db1 fix
2 months ago
wangbluo 65c8297710 fix the attn
2 months ago
wangbluo cfd9eda628 fix the ring attn
2 months ago
binmakeswell cbaa104216
release FP8 news (#6068)
2 months ago
Hongxin Liu dabc2e7430
[release] update version (#6062)
2 months ago
Camille Zhong f9546ba0be
[ColossalEval] support for vllm (#6056)
2 months ago
botbw 4fa6b9509c
[moe] add parallel strategy for shared_expert && fix test for deepseek (#6063)
2 months ago
Wang Binluo 63314ce4e4
Merge pull request #6064 from wangbluo/fix_attn
2 months ago
wangbluo 10e4f7da72 fix
2 months ago
Wang Binluo 37e35230ff
Merge pull request #6061 from wangbluo/sp_fix
2 months ago
wangbluo 827ef3ee9a fix
2 months ago
Guangyao Zhang bdb125f83f
[doc] FP8 training and communication document (#6050)
2 months ago
Guangyao Zhang f20b066c59
[fp8] Disable all_gather intranode. Disable Redundant all_gather fp8 (#6059)
2 months ago
wangbluo b582319273 fix
2 months ago
wangbluo 0ad3129cb9 fix
2 months ago
wangbluo 0b14a5512e fix
2 months ago
botbw 696fced0d7
[fp8] fix missing fp8_comm flag in mixtral (#6057)
2 months ago
wangbluo dc032172c3 fix
2 months ago
wangbluo f393867cff fix
2 months ago
wangbluo 6eb8832366 fix
2 months ago
wangbluo 683179cefd fix
2 months ago
wangbluo 0a01e2a453 fix the attn
2 months ago
pre-commit-ci[bot] 216d54e374 [pre-commit.ci] auto fixes from pre-commit.com hooks
2 months ago
wangbluo fdd84b9087 fix the sp
2 months ago
flybird11111 a35a078f08
[doc] update sp doc (#6055)
2 months ago
Hongxin Liu 13946c4448
[fp8] hotfix backward hook (#6053)
2 months ago
botbw c54c4fcd15
[hotfix] moe hybrid parallelism benchmark & follow-up fix (#6048)
2 months ago
Wenxuan Tan 8fd25d6e09
[Feature] Split cross-entropy computation in SP (#5959)
3 months ago
Hongxin Liu b3db1058ec
[release] update version (#6041)
3 months ago
Hanks 5ce6dd75bf
[fp8] disable all_to_all_fp8 in intranode (#6045)
3 months ago
Hongxin Liu 26e553937b
[fp8] fix linear hook (#6046)
3 months ago
Hongxin Liu c3b5caff0e
[fp8] optimize all-gather (#6043)
3 months ago
Tong Li c650a906db
[Hotfix] Remove deprecated install (#6042)
3 months ago
Gao, Ruiyuan e9032fb0b2
[colossalai/checkpoint_io/...] fix bug in load_state_dict_into_model; format error msg (#6020)
3 months ago
Guangyao Zhang e96a0761ea
[FP8] unsqueeze scale to make it compatible with torch.compile (#6040)
3 months ago
Tong Li 0d3a85d04f
add fused norm (#6038)
3 months ago
Tong Li 4a68efb7da
[Colossal-LLaMA] Refactor latest APIs (#6030)
3 months ago
Hongxin Liu cc1b0efc17
[plugin] hotfix zero plugin (#6036)
3 months ago
Wenxuan Tan d383449fc4
[CI] Remove triton version for compatibility bug; update req torch >=2.2 (#6018)
3 months ago
Hongxin Liu 17904cb5bf
Merge pull request #6012 from hpcaitech/feature/fp8_comm
3 months ago
Wang Binluo 4a6f31eb0c
Merge pull request #6033 from wangbluo/fix
3 months ago