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.

botbw 2431694564 [moe] implement transit between non moe tp and ep		4 months ago
..
chatglm2_6b	[pre-commit.ci] pre-commit autoupdate (#5572)	5 months ago
__init__.py	[shardformer] added development protocol for standardization (#4149)	1 year ago
bert.py	[shardformer]delete xformers (#5859)	5 months ago
blip2.py	[shardformer] support bias_gelu_jit_fused for models (#5647)	7 months ago
bloom.py	[Feature] Enable PP + SP for llama (#5868)	5 months ago
chatglm2.py	[ShardFormer] Add Ulysses Sequence Parallelism support for Command-R, Qwen2 and ChatGLM (#5897)	5 months ago
command.py	[ShardFormer] Add Ulysses Sequence Parallelism support for Command-R, Qwen2 and ChatGLM (#5897)	5 months ago
deepseek.py	[shardformer] DeepseekMoE support (#5871)	5 months ago
falcon.py	[shardformer] fix modeling of bloom and falcon (#5796)	6 months ago
gpt2.py	[Feature] Enable PP + SP for llama (#5868)	5 months ago
gptj.py	[shardformer] upgrade transformers to 4.39.3 (#5815)	6 months ago
jit.py	[misc] update pre-commit and run all files (#4752)	1 year ago
llama.py	[Feature] Enable PP + SP for llama (#5868)	5 months ago
mistral.py	[Feature] Enable PP + SP for llama (#5868)	5 months ago
mixtral.py	[moe] implement transit between non moe tp and ep	4 months ago
opt.py	[Feature] Enable PP + SP for llama (#5868)	5 months ago
qwen2.py	[ShardFormer] fix qwen2 sp (#5903)	5 months ago
sam.py	[shardformer]delete xformers (#5859)	5 months ago
t5.py	[shardformer] Support the T5ForTokenClassification model (#5816)	5 months ago
vit.py	[shardformer] support bias_gelu_jit_fused for models (#5647)	7 months ago
whisper.py	[shardformer] upgrade transformers to 4.39.3 (#5815)	6 months ago