github
/
ColossalAI
mirror of https://github.com/hpcaitech/ColossalAI
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
Making large AI models cheaper, faster and more accessible
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.
3189 Commits
31 Branches
46 Tags
236 MiB
Branch: feat/speculative-decoding
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'feat/speculative-decoding'
${ noResults }
ColossalAI/extensions/csrc/__init__.py

12 lines
350 B
Raw Permalink Normal View History Unescape

add colossalai kernel module (#55)
3 years ago
from .layer_norm import MixedFusedLayerNorm as LayerNorm
refactor kernel (#142)
3 years ago
from .multihead_attention import MultiHeadAttention
[shardformer] update shardformer to use flash attention 2 (#4392) * cherry-pick flash attention 2 cherry-pick flash attention 2 * [shardformer] update shardformer to use flash attention 2 [shardformer] update shardformer to use flash attention 2, fix [shardformer] update shardformer to use flash attention 2, fix [shardformer] update shardformer to use flash attention 2, fix
1 year ago
from .scaled_softmax import AttnMaskType, FusedScaleMaskSoftmax, ScaledUpperTriangMaskedSoftmax
[kernel] fixed repeated loading of kernels (#2549) * [kernel] fixed repeated loading of kernels * polish code * polish code
2 years ago
[fix] coloattention support flash attention 2 (#4347) Improved ColoAttention interface to support flash attention 2. Solved #4322
1 year ago
__all__ = [
[misc] update pre-commit and run all files (#4752) * [misc] update pre-commit * [misc] run pre-commit * [misc] remove useless configuration files * [misc] ignore cuda for clang-format
1 year ago
"LayerNorm",
"MultiHeadAttention",
"FusedScaleMaskSoftmax",
"ScaledUpperTriangMaskedSoftmax",
"AttnMaskType",
[fix] coloattention support flash attention 2 (#4347) Improved ColoAttention interface to support flash attention 2. Solved #4322
1 year ago
]