github
/
ColossalAI
mirror of https://github.com/hpcaitech/ColossalAI
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
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.
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '8e08c27e19'
${ noResults }
ColossalAI/colossalai/shardformer/policies
History
Hongxin Liu dc2cdaf3e8
[shardformer] optimize seq parallelism (#6086) 		2 months ago
..
__init__.py
…
auto_policy.py
…
base_policy.py [fp8] Merge feature/fp8_comm to main branch of Colossalai (#6016) 3 months ago
bert.py [shardformer] optimize seq parallelism (#6086) 2 months ago
blip2.py [shardformer] fix linear 1d row and support uneven splits for fused qkv linear (#6084) 2 months ago
bloom.py [shardformer] optimize seq parallelism (#6086) 2 months ago
chatglm2.py [shardformer] optimize seq parallelism (#6086) 2 months ago
command.py [fp8] Merge feature/fp8_comm to main branch of Colossalai (#6016) 3 months ago
deepseek.py [hotfix] moe hybrid parallelism benchmark & follow-up fix (#6048) 3 months ago
falcon.py [fp8] support hybrid parallel plugin (#5982) 3 months ago
gpt2.py [shardformer] optimize seq parallelism (#6086) 2 months ago
gptj.py [shardformer] optimize seq parallelism (#6086) 2 months ago
llama.py [Feature] Split cross-entropy computation in SP (#5959) 3 months ago
mistral.py [fp8] Merge feature/fp8_comm to main branch of Colossalai (#6016) 3 months ago
mixtral.py [fp8] fix missing fp8_comm flag in mixtral (#6057) 2 months ago
opt.py [fp8] support hybrid parallel plugin (#5982) 3 months ago
qwen2.py [Feature] Zigzag Ring attention (#5905) 3 months ago
sam.py [shardformer] fix linear 1d row and support uneven splits for fused qkv linear (#6084) 2 months ago
t5.py [fp8] support hybrid parallel plugin (#5982) 3 months ago
vit.py [fp8] support hybrid parallel plugin (#5982) 3 months ago
whisper.py [fp8] support hybrid parallel plugin (#5982) 3 months ago