You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
flybird11111
451e9142b8
fix flash attn (#5209)
|
11 months ago |
| .. |
|
__init__.py
|
[shardformer] init shardformer code structure (#3731)
|
1 year ago |
|
auto_policy.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
base_policy.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
bert.py
|
[pipeline]: fix p2p comm, add metadata cache and support llama interleaved pp (#5134)
|
11 months ago |
|
blip2.py
|
[hotfix] Add layer norm gradients all-reduce for sequence parallel (#4926)
|
1 year ago |
|
bloom.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
chatglm2.py
|
[Inference] Fix bug in ChatGLM2 Tensor Parallelism (#5014)
|
1 year ago |
|
falcon.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
gpt2.py
|
[hotfix] Add layer norm gradients all-reduce for sequence parallel (#4926)
|
1 year ago |
|
gptj.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
llama.py
|
fix flash attn (#5209)
|
11 months ago |
|
mistral.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
opt.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
|
sam.py
|
[hotfix] Add layer norm gradients all-reduce for sequence parallel (#4926)
|
1 year ago |
|
t5.py
|
[gemini] gemini support tensor parallelism. (#4942)
|
1 year ago |
|
vit.py
|
[hotfix] Add layer norm gradients all-reduce for sequence parallel (#4926)
|
1 year ago |
|
whisper.py
|
[shardformer]: support gpt-j, falcon, Mistral and add interleaved pipeline for bert (#5088)
|
1 year ago |
