liger-kernel-nightly 0.5.5.dev20250402185606__py3-none-any.whl → 0.5.5.dev20250402212634__py3-none-any.whl

liger_kernel/chunked_loss/fused_linear_ppo.py

@@ -37,6 +37,7 @@ class LigerFusedLinearPPOBase(torch.autograd.Function):
         use_ref_model=False,
         chunk_size=1,
     ):
+        # TODO: check torch compile matmul
         """Chunked forward pass for PPO loss computation.
 
         Args:

liger_kernel/ops/fused_linear_jsd.py

@@ -8,11 +8,12 @@ from liger_kernel.ops.utils import amp_custom_bwd
 from liger_kernel.ops.utils import amp_custom_fwd
 from liger_kernel.ops.utils import element_mul_kernel
 from liger_kernel.ops.utils import is_hip
+from liger_kernel.utils import infer_device
 
 # 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
+MAX_FUSED_SIZE = 4096 if infer_device() == "xpu" else 65536 // 2
 
 
 def fused_linear_jsd_forward(

{liger_kernel_nightly-0.5.5.dev20250402185606.dist-info → liger_kernel_nightly-0.5.5.dev20250402212634.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: liger_kernel_nightly
-Version: 0.5.5.dev20250402185606
+Version: 0.5.5.dev20250402212634
 Summary: Efficient Triton kernels for LLM Training
 License: BSD 2-CLAUSE LICENSE
         Copyright 2024 LinkedIn Corporation 

{liger_kernel_nightly-0.5.5.dev20250402185606.dist-info → liger_kernel_nightly-0.5.5.dev20250402212634.dist-info}/RECORD

@@ -7,7 +7,7 @@ liger_kernel/chunked_loss/cpo_loss.py,sha256=Gzz1eU4kgcbdubFVRy55e8A1Cr-r45UgNic
 liger_kernel/chunked_loss/dpo_loss.py,sha256=xZwGqS04si9zXyob95SAdalC-hajZg8fWINqiqffN8k,5855
 liger_kernel/chunked_loss/functional.py,sha256=9G3nKm-Bi7uoZRFkL8wwGMl6juDl4bSzDvTa5GHZPzg,955
 liger_kernel/chunked_loss/fused_linear_distillation.py,sha256=ooR-qnZCyWJN935oHCSWLaKKKyaYERyhNczRGi1VOiw,11935
-liger_kernel/chunked_loss/fused_linear_ppo.py,sha256=2_UvvIksUP45RBw3c-88-jOtjGATf04vaWopcqtX4Oo,12688
+liger_kernel/chunked_loss/fused_linear_ppo.py,sha256=-E4AuWY-y2bMo_kAmEQBgQ92UJh3L5IiCRGVcfMJOCE,12731
 liger_kernel/chunked_loss/fused_linear_preference.py,sha256=ojB42jYPu0c4ki96Ft-hy7Sf6fh_WikG-aWNrlZzSio,18362
 liger_kernel/chunked_loss/fused_linear_unpaired_preference.py,sha256=RiuK3UtRwH9T6jZ36sA8Urj-TVuOLOO2syLg_JOQapY,13437
 liger_kernel/chunked_loss/grpo_loss.py,sha256=6Mb4ZT6MfnOr4Xo681rMR0LKkhzJhInvQp8wp2YVMK0,8913
@@ -19,7 +19,7 @@ liger_kernel/ops/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,
 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
+liger_kernel/ops/fused_linear_jsd.py,sha256=CSoprxb-YcJy-YUKiTcYkxN8sb9h2kdk_iHuncvSV5c,9683
 liger_kernel/ops/geglu.py,sha256=axGvCIvlBzuluoAIrWTsp2iZM4BFKNInkPov8YVvH9E,4126
 liger_kernel/ops/group_norm.py,sha256=qD4D4lSjSgVtO52EBNLC2iTseALRgPgqXE50U2woggk,10837
 liger_kernel/ops/jsd.py,sha256=rkloGA7nDfVaa5nKY6-EYBw0E1p_MSsl4fr2xZGTp04,6961
@@ -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.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,,
+liger_kernel_nightly-0.5.5.dev20250402212634.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
+liger_kernel_nightly-0.5.5.dev20250402212634.dist-info/METADATA,sha256=PaFO566AhWjPHX3kn2S83vBHlK0N6LgyYjXL8SvH2qs,22959
+liger_kernel_nightly-0.5.5.dev20250402212634.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
+liger_kernel_nightly-0.5.5.dev20250402212634.dist-info/WHEEL,sha256=iAkIy5fosb7FzIOwONchHf19Qu7_1wCWyFNR5gu9nU0,91
+liger_kernel_nightly-0.5.5.dev20250402212634.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
+liger_kernel_nightly-0.5.5.dev20250402212634.dist-info/RECORD,,