PanQiWei 26dc6852fe support inherit one of the three fused attention class and customize attn_bias building logic		2023-08-07 18:59:04 +08:00
..
fused_modules	support inherit one of the three fused attention class and customize attn_bias building logic	2023-08-07 18:59:04 +08:00
qlinear	remove unnecessary lines	2023-08-06 16:24:44 +08:00
triton_utils	support 32dim triton kernel	2023-06-02 19:04:12 +09:00
__init__.py	refactor file structure	2023-04-25 18:58:20 +08:00
_fused_base.py	add trainable mode	2023-05-26 13:11:30 +08:00
fused_gptj_attn.py	add trainable mode	2023-05-26 13:11:30 +08:00
fused_llama_attn.py	add trainable mode	2023-05-26 13:11:30 +08:00
fused_llama_mlp.py	update FusedLlamaMLPForQuantizedModel for general usage purpose	2023-05-27 07:47:20 +08:00