A

apple/axlearn

Settings | Report Duplicate

0

I Use This!

High Activity

Commits : Listings

Analyzed about 9 hours ago. based on code collected about 9 hours ago.

Commit Message	Contributor	Files Modified	Lines Added	Lines Removed	Code Location	Date
Aug 23, 2024 — Aug 23, 2025 Showing page 14 of 38 Search / Filter on:
Introduce the scale enum flag in Embedding layer for LLM embedding. (#909)	Dongseong Hwang	More...				8 months ago
Flash2 and supports cross attention and dropout (#905)	Hanzhi Zhou	More...				8 months ago
Allow parallel gpu tests (#910)	Hanzhi Zhou	More...				8 months ago
Optimize TPU Flash Attention (20x XLA compilation speed-up on 32k long context) (#908)	Dongseong Hwang	More...				8 months ago
The codebook of the KmeansVectorQuantizer should be initialized with scale=1/sqrt(dim). (#907)	Dongseong Hwang	More...				8 months ago
`AdaptiveLayerNormModulation` now supports sequence conditions. (#906)	Dongseong Hwang	More...				8 months ago
Implements FlashDecoding with Sparsity Support (#899)	Hanzhi Zhou	More...				8 months ago
Fix softmax scale (#903)	Hanzhi Zhou	More...				8 months ago
Little clean-up in frontend. (#901)	Dongseong Hwang	More...				8 months ago
Learner: add new unittests using Model. (#900)	Dongseong Hwang	More...				8 months ago
Renames sm_scale to softmax_scale for consistency. (#894)	Ruoming Pang	More...				8 months ago
MaskFnAttentionBias._bool_value passes the same rank position tensors to mask_fn. (#888)	Dongseong Hwang	More...				8 months ago
Adds @apple/axlearn-admins to CODEOWNERS. (#897)	tgunter	More...				8 months ago
Cleanup `MaskFnAttentionBias.target_positions`. (#895)	apghml	More...				8 months ago
Update RoformerQKVLinear to support kv_state (#882)	qdavid1	More...				8 months ago
Support fine grained activation sharding. (#21) (#881)	Patrick Toulme	More...				8 months ago
Fix v6e boolean flags (#889)	Meng (Ethan) Li	More...				8 months ago
Add default compiler options for v6e (#887)	Sam Stoelinga	More...				8 months ago
Support v6e (#879)	Meng (Ethan) Li	More...				9 months ago
fix issues with tool use metrics (#878)	fnan	More...				9 months ago
Makes Bastion write scheduler history files (#877)	Ruoming Pang	More...				9 months ago
DiT: implement init_states and extend_step for DiT transformer (#876)	Dongseong Hwang	More...				9 months ago
pin scikit-learn (#875)	fnan	More...				9 months ago
Simplifies ScheduleResults.job_verdicts (#872)	Ruoming Pang	More...				9 months ago
Refactor attention bias/mask/segments. (#873)	apghml	More...				9 months ago
add tool use intent metrics (#871)	fnan	More...				9 months ago
Support orbax state builder (#866)	Hanzhi Zhou	More...				9 months ago
add merge_group to action workflows (#870)	Mike Drob	More...				9 months ago
Refactoring Convolution; factor out convolutions from layers.py (#865)	Dongseong Hwang	More...				9 months ago
Fix feedforward remat point (#864)	Hanzhi Zhou	More...				9 months ago

←
1
2
…
10
11
12
13
14
15
16
17
18
…
37
38
→