Dan Saunders d4e29e5b67 adding split_heads argument for retaining original (Q, K) dimensionanlity

2024-12-20 20:39:53 +00:00

..

convert_differential_transformer

adding split_heads argument for retaining original (Q, K) dimensionanlity

2024-12-20 20:39:53 +00:00

__init__.py

add e2e smoke tests for llama liger integration (#1884)

2024-09-01 19:29:37 -04:00

liger.py

feat: add cut_cross_entropy (#2091)

2024-12-03 08:22:22 -05:00

test_cut_cross_entropy.py

need to update deepspeed version in extras too (#2161) [skip ci]

2024-12-09 14:01:44 -05:00