liger-kernel-nightly 0.5.3.dev20250221002845__py3-none-any.whl → 0.5.3.dev20250221003838__py3-none-any.whl

Sign up to get free protection for your applications and to get access to all the features.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: liger_kernel_nightly
3
- Version: 0.5.3.dev20250221002845
3
+ Version: 0.5.3.dev20250221003838
4
4
  Summary: Efficient Triton kernels for LLM Training
5
5
  License: BSD 2-CLAUSE LICENSE
6
6
  Copyright 2024 LinkedIn Corporation
@@ -126,7 +126,7 @@ Requires-Dist: mkdocs-material; extra == "dev"
126
126
 
127
127
  **Liger Kernel** is a collection of Triton kernels designed specifically for LLM training. It can effectively increase multi-GPU **training throughput by 20%** and reduces **memory usage by 60%**. We have implemented **Hugging Face Compatible** `RMSNorm`, `RoPE`, `SwiGLU`, `CrossEntropy`, `FusedLinearCrossEntropy`, and more to come. The kernel works out of the box with [Flash Attention](https://github.com/Dao-AILab/flash-attention), [PyTorch FSDP](https://pytorch.org/tutorials/intermediate/FSDP_tutorial.html), and [Microsoft DeepSpeed](https://github.com/microsoft/DeepSpeed). We welcome contributions from the community to gather the best kernels for LLM training.
128
128
 
129
- We've also added optimized Post-Training kernels that deliver **up to 80% memory savings** for alignment and distillation tasks. We support losses like DPO, CPO, ORPO, SimPO, JSD, and many more. Check out [how we optimize the memory](https://x.com/hsu_byron/status/1866577403918917655).
129
+ We've also added optimized Post-Training kernels that deliver **up to 80% memory savings** for alignment and distillation tasks. We support losses like DPO, CPO, ORPO, SimPO, KTO, JSD, and many more. Check out [how we optimize the memory](https://x.com/hsu_byron/status/1866577403918917655).
130
130
 
131
131
  ## Supercharge Your Model with Liger Kernel
132
132
 
@@ -341,6 +341,7 @@ loss.backward()
341
341
  | Fused Linear DPO Loss | `liger_kernel.chunked_loss.LigerFusedLinearDPOLoss` |
342
342
  | Fused Linear ORPO Loss | `liger_kernel.chunked_loss.LigerFusedLinearORPOLoss` |
343
343
  | Fused Linear SimPO Loss | `liger_kernel.chunked_loss.LigerFusedLinearSimPOLoss` |
344
+ | Fused Linear KTO Loss | `liger_kernel.chunked_loss.LigerFusedLinearKTOLoss` |
344
345
 
345
346
  ### Distillation Kernels
346
347
 
@@ -65,9 +65,9 @@ liger_kernel/transformers/trainer/__init__.py,sha256=p7yQfklV8-467qSz_ZMimkbDF7H
65
65
  liger_kernel/transformers/trainer/orpo_trainer.py,sha256=pdekW7l6Qg_aqa5SYKYlSWUF8m3lkOFvFLcIMEHrz9s,8338
66
66
  liger_kernel/triton/__init__.py,sha256=qCiCamzCRv6lpV8IqpAc9YMdNKC7GKurClWceQPnlis,92
67
67
  liger_kernel/triton/monkey_patch.py,sha256=Rd0hUHAzDkFfHvnX7-PBaNK5EKnZhtfM_h-fgQH9HPY,1568
68
- liger_kernel_nightly-0.5.3.dev20250221002845.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
69
- liger_kernel_nightly-0.5.3.dev20250221002845.dist-info/METADATA,sha256=IppIvVS1LGihuHadPcZP-l4F7Q5bZH4C5-DuEVlYjqo,21864
70
- liger_kernel_nightly-0.5.3.dev20250221002845.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
71
- liger_kernel_nightly-0.5.3.dev20250221002845.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
72
- liger_kernel_nightly-0.5.3.dev20250221002845.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
73
- liger_kernel_nightly-0.5.3.dev20250221002845.dist-info/RECORD,,
68
+ liger_kernel_nightly-0.5.3.dev20250221003838.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
69
+ liger_kernel_nightly-0.5.3.dev20250221003838.dist-info/METADATA,sha256=zJp1YMQDbbOzeNRKFf7AN7hYqePdT49OEMQkN_buKl8,21963
70
+ liger_kernel_nightly-0.5.3.dev20250221003838.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
71
+ liger_kernel_nightly-0.5.3.dev20250221003838.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
72
+ liger_kernel_nightly-0.5.3.dev20250221003838.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
73
+ liger_kernel_nightly-0.5.3.dev20250221003838.dist-info/RECORD,,