liger-kernel-nightly 0.5.5.dev20250402184001__py3-none-any.whl → 0.5.5.dev20250402185606__py3-none-any.whl

liger_kernel/ops/cross_entropy.py

@@ -9,6 +9,7 @@ import triton.language as tl
 from liger_kernel.ops.utils import compare_version
 from liger_kernel.ops.utils import element_mul_kernel
 from liger_kernel.ops.utils import is_hip
+from liger_kernel.utils import infer_device
 
 if compare_version("triton", operator.ge, "3.0.0"):
     try:
@@ -59,7 +60,7 @@ def liger_cross_entropy_kernel(
     z_loss_ptr: Pointer to tensor to store the z loss. No operation if RETURN_Z_LOSS is 0.
     loss_stride (int): The stride of the loss tensor.
     n_cols (int): The number of columns in the input tensor.
-    n_non_ignore (flaot): The number of non-ignored elements in the batch.
+    n_non_ignore (float): The number of non-ignored elements in the batch.
     sum_non_ignore_weight (float): The sum of non-ignored target's weights in the batch.
     weight_sum (float): The sum of weight tensor.
     ignore_index (int): The index to ignore in the target.
@@ -258,7 +259,7 @@ def liger_cross_entropy_kernel(
 # The hard limit of TRITON_MAX_TENSOR_NUMEL is 1048576 https://github.com/triton-lang/triton/blob/ba42a5c68fd0505f8c42f4202d53be0f8d9a5fe0/python/triton/language/core.py#L19
 # However, setting limit as 65536 as in LayerNorm tutorial is faster because of less register spilling
 # The optimal maximum block size depends on your hardware, your kernel, and your dtype
-MAX_FUSED_SIZE = 65536 // 2  # the best size we found by manually tuning
+MAX_FUSED_SIZE = 4096 if infer_device() == "xpu" else 65536 // 2  # the best size we found by manually tuning
 
 
 def cross_entropy_forward(

{liger_kernel_nightly-0.5.5.dev20250402184001.dist-info → liger_kernel_nightly-0.5.5.dev20250402185606.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: liger_kernel_nightly
-Version: 0.5.5.dev20250402184001
+Version: 0.5.5.dev20250402185606
 Summary: Efficient Triton kernels for LLM Training
 License: BSD 2-CLAUSE LICENSE
         Copyright 2024 LinkedIn Corporation 

{liger_kernel_nightly-0.5.5.dev20250402184001.dist-info → liger_kernel_nightly-0.5.5.dev20250402185606.dist-info}/RECORD

@@ -16,7 +16,7 @@ liger_kernel/chunked_loss/kto_loss.py,sha256=llVCe6DkcpCo57seGWoMikaQVFApx764jsm
 liger_kernel/chunked_loss/orpo_loss.py,sha256=nu9UYG16dcMw93lvHi4_hYs3Q0FK1KnlmMRj7OpYU8s,4872
 liger_kernel/chunked_loss/simpo_loss.py,sha256=fy2w8KbhMrBv7b1jdIeH3bBFxY52bPQPZb3KwBvmurM,5385
 liger_kernel/ops/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-liger_kernel/ops/cross_entropy.py,sha256=yKKhN63I7r9NxJye4wTLBvvKAyrXQt6jf4nBo3lJyVg,18860
+liger_kernel/ops/cross_entropy.py,sha256=T5oSsqOS1y-Iea5o9v_BSU-_mIEXqWAT1oX_m59NcA4,18941
 liger_kernel/ops/dyt.py,sha256=YD1-buHz9VmIX838VKzLc-lm5CeUQ4LAskGDWBUMQHA,6187
 liger_kernel/ops/fused_linear_cross_entropy.py,sha256=1Y3Uk_TCSjqKgoG2eot1ptnWXJXXQESqGvOmqAW1gsM,10912
 liger_kernel/ops/fused_linear_jsd.py,sha256=Seshez2qaM6HiTQ8_HEqSwhaeVruNT1SvIM4ZrAPBEU,9602
@@ -72,9 +72,9 @@ liger_kernel/transformers/trainer/__init__.py,sha256=p7yQfklV8-467qSz_ZMimkbDF7H
 liger_kernel/transformers/trainer/orpo_trainer.py,sha256=pdekW7l6Qg_aqa5SYKYlSWUF8m3lkOFvFLcIMEHrz9s,8338
 liger_kernel/triton/__init__.py,sha256=qCiCamzCRv6lpV8IqpAc9YMdNKC7GKurClWceQPnlis,92
 liger_kernel/triton/monkey_patch.py,sha256=Rd0hUHAzDkFfHvnX7-PBaNK5EKnZhtfM_h-fgQH9HPY,1568
-liger_kernel_nightly-0.5.5.dev20250402184001.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
-liger_kernel_nightly-0.5.5.dev20250402184001.dist-info/METADATA,sha256=DLGGPCgn1-dKSQVP5sYIzzRoh7c9wBUjM7JFujYn1KI,22959
-liger_kernel_nightly-0.5.5.dev20250402184001.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
-liger_kernel_nightly-0.5.5.dev20250402184001.dist-info/WHEEL,sha256=iAkIy5fosb7FzIOwONchHf19Qu7_1wCWyFNR5gu9nU0,91
-liger_kernel_nightly-0.5.5.dev20250402184001.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
-liger_kernel_nightly-0.5.5.dev20250402184001.dist-info/RECORD,,
+liger_kernel_nightly-0.5.5.dev20250402185606.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
+liger_kernel_nightly-0.5.5.dev20250402185606.dist-info/METADATA,sha256=XQaGc9bnsEFdwtLh1Mv5_fX-TIejLbcHk1SP-FEY5ew,22959
+liger_kernel_nightly-0.5.5.dev20250402185606.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
+liger_kernel_nightly-0.5.5.dev20250402185606.dist-info/WHEEL,sha256=iAkIy5fosb7FzIOwONchHf19Qu7_1wCWyFNR5gu9nU0,91
+liger_kernel_nightly-0.5.5.dev20250402185606.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
+liger_kernel_nightly-0.5.5.dev20250402185606.dist-info/RECORD,,