Merge branch 'main' into Dao-AILab/main

26f4b5fb · Woosuk Kwon · 5018ac6a · 12375706 · 5018ac6a · 5018ac6a
Commit 26f4b5fb authored Jul 31, 2024 by Woosuk Kwon
20 changed files
--- a/flash_attn/models/btlm.py
+++ b/flash_attn/models/btlm.py
--- a/flash_attn/models/falcon.py
+++ b/flash_attn/models/falcon.py
--- a/flash_attn/models/gpt.py
+++ b/flash_attn/models/gpt.py
--- a/flash_attn/models/gpt_neox.py
+++ b/flash_attn/models/gpt_neox.py
--- a/flash_attn/models/gptj.py
+++ b/flash_attn/models/gptj.py
--- a/flash_attn/models/llama.py
+++ b/flash_attn/models/llama.py
--- a/flash_attn/models/opt.py
+++ b/flash_attn/models/opt.py
--- a/flash_attn/models/vit.py
+++ b/flash_attn/models/vit.py
--- a/flash_attn/modules/__init__.py
+++ b/flash_attn/modules/__init__.py
--- a/flash_attn/modules/block.py
+++ b/flash_attn/modules/block.py
--- a/flash_attn/modules/embedding.py
+++ b/flash_attn/modules/embedding.py
--- a/flash_attn/modules/mha.py
+++ b/flash_attn/modules/mha.py
--- a/flash_attn/modules/mlp.py
+++ b/flash_attn/modules/mlp.py
--- a/flash_attn/ops/__init__.py
+++ b/flash_attn/ops/__init__.py
--- a/flash_attn/ops/activations.py
+++ b/flash_attn/ops/activations.py
--- a/flash_attn/ops/fused_dense.py
+++ b/flash_attn/ops/fused_dense.py
--- a/flash_attn/ops/layer_norm.py
+++ b/flash_attn/ops/layer_norm.py
--- a/flash_attn/ops/rms_norm.py
+++ b/flash_attn/ops/rms_norm.py
--- a/flash_attn/ops/triton/__init__.py
+++ b/flash_attn/ops/triton/__init__.py
--- a/flash_attn/ops/triton/cross_entropy.py
+++ b/flash_attn/ops/triton/cross_entropy.py