 AlpinDale
|
cbb90f0b2a
add safetensor support
|
1 year ago |
|
AlpinDale
|
fffb9f2dac
chore: attention kernel optimizations
|
1 year ago |
|
AlpinDale
|
acbf49ef89
feat: refactor scheduler
|
1 year ago |
|
AlpinDale
|
10334ebd7c
requirements: stricter version for fschat
|
1 year ago |
|
AlpinDale
|
bc725a82e8
fix: leftover name changes from an old migration
|
1 year ago |
|
AlpinDale
|
592ee204a6
fix: ray depends on pyarrow
|
1 year ago |
|
AlpinDale
|
4472e432cf
fix: requirements again.
|
1 year ago |
|
AlpinDale
|
7991c14e51
fix: requirements and accidental commit
|
1 year ago |
|
AlpinDale
|
4ac5560152
fix: revert template change for now
|
1 year ago |
|
AlpinDale
|
bf132e29d6
feat: bump up version
|
1 year ago |
 AlpinDale
|
76b2e4a445
Merge dev branch into main (#7)
|
1 year ago |
|
AlpinDale
|
3cdff3cd8b
readme: add common issues
|
1 year ago |
|
AlpinDale
|
e8eac42213
fix: incorrect call
|
1 year ago |
|
AlpinDale
|
b188d1093b
test: throughput
|
1 year ago |
|
AlpinDale
|
c10b83422d
readme: add notes
|
1 year ago |
|
AlpinDale
|
908091008e
readme: typo
|
1 year ago |
|
AlpinDale
|
229733d39f
feat: bump up the version
|
1 year ago |
|
AlpinDale
|
6dfca19dda
fix: gpt-j loading
|
1 year ago |
|
AlpinDale
|
d6f705d90b
readme: more info on usage
|
1 year ago |
|
AlpinDale
|
89c7f0469f
fix: calculate the key/value outputs with kvhead
|
1 year ago |
|
AlpinDale
|
06e71fc492
feat: add GPT-NeoX support for testing purposes
|
1 year ago |
|
AlpinDale
|
38e7b98bec
readme: typo
|
1 year ago |
|
AlpinDale
|
751f759649
readme: more detailed install instructions
|
1 year ago |
|
AlpinDale
|
97bb098066
fix: typo lol
|
1 year ago |
|
AlpinDale
|
14e73807e6
fix: typo
|
1 year ago |
|
AlpinDale
|
35ec43f478
fix: remove aria2 for now
|
1 year ago |
|
AlpinDale
|
24c78e7306
optimization: multi-query attention kernel
|
1 year ago |
|
AlpinDale
|
6de30f43a4
fix: typos and refactors for llama
|
1 year ago |
|
AlpinDale
|
cd17af7bc4
chore: clean up gpt-j and remove cross_attention
|
1 year ago |
|
AlpinDale
|
76162fe366
chore: optimize sampler for gpt-j
|
1 year ago |
