first commit

34e67b1e · zhangshao · 34e67b1e · 34e67b1e · 34e67b1e · 34e67b1e
Commit 34e67b1e authored May 09, 2026 by zhangshao
20 changed files
--- a/flash_attn/ops/triton/cross_entropy.py
+++ b/flash_attn/ops/triton/cross_entropy.py
--- a/flash_attn/ops/triton/k_activations.py
+++ b/flash_attn/ops/triton/k_activations.py
--- a/flash_attn/ops/triton/layer_norm.py
+++ b/flash_attn/ops/triton/layer_norm.py
--- a/flash_attn/ops/triton/linear.py
+++ b/flash_attn/ops/triton/linear.py
--- a/flash_attn/ops/triton/mlp.py
+++ b/flash_attn/ops/triton/mlp.py
--- a/flash_attn/ops/triton/rotary.py
+++ b/flash_attn/ops/triton/rotary.py
--- a/flash_attn/pyproject.toml
+++ b/flash_attn/pyproject.toml
--- a/flash_attn/utils/__init__.py
+++ b/flash_attn/utils/__init__.py
--- a/flash_attn/utils/benchmark.py
+++ b/flash_attn/utils/benchmark.py
--- a/flash_attn/utils/distributed.py
+++ b/flash_attn/utils/distributed.py
--- a/flash_attn/utils/generation.py
+++ b/flash_attn/utils/generation.py
--- a/flash_attn/utils/pretrained.py
+++ b/flash_attn/utils/pretrained.py
--- a/flash_attn/utils/sparse_utils.py
+++ b/flash_attn/utils/sparse_utils.py
--- a/get_version.py
+++ b/get_version.py
--- a/setup.py
+++ b/setup.py
--- a/tests/hy/fa_fwd_padding_mask.py
+++ b/tests/hy/fa_fwd_padding_mask.py
--- a/tests/layers/test_rotary.py
+++ b/tests/layers/test_rotary.py
--- a/tests/losses/test_cross_entropy.py
+++ b/tests/losses/test_cross_entropy.py
--- a/tests/losses/test_cross_entropy_parallel.py
+++ b/tests/losses/test_cross_entropy_parallel.py
--- a/tests/models/test_baichuan.py
+++ b/tests/models/test_baichuan.py