Liger-kernel

What's Changed
* Infer correct device for AMD HIP device by helloworld1 in https://github.com/linkedin/Liger-Kernel/pull/587
* add out of bounds check to cross entropy by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/588
* Monkeypatch for Qwen2.5-VL by BenasdTW in https://github.com/linkedin/Liger-Kernel/pull/552
* KTO changes to return aux outputs by vaibhavjindal in https://github.com/linkedin/Liger-Kernel/pull/589
* [KTO] Only return summed metrics by vaibhavjindal in https://github.com/linkedin/Liger-Kernel/pull/591
* increase chunk size for distillation and add bias to jsd by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/590
* [CI] Add ROCm 6.3 CI by tjtanaa in https://github.com/linkedin/Liger-Kernel/pull/506
* Fix KTO speed issue by vaibhavjindal in https://github.com/linkedin/Liger-Kernel/pull/592
* Compare means of aggregated outputs in KTO tests by vaibhavjindal in https://github.com/linkedin/Liger-Kernel/pull/595
* Fix means of logps and rewards by vaibhavjindal in https://github.com/linkedin/Liger-Kernel/pull/597
* Add chunk_size param to chunked losses by RichhLi in https://github.com/linkedin/Liger-Kernel/pull/599
* Fix DPO/ORPO typo in readme by tyler-romero in https://github.com/linkedin/Liger-Kernel/pull/602
* version bump by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/605

New Contributors
* RichhLi made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/599

**Full Changelog**: https://github.com/linkedin/Liger-Kernel/compare/v0.5.4...v0.5.5

What's Changed
* add GitHub CI for Intel GPU by faaany in https://github.com/linkedin/Liger-Kernel/pull/536
* Add Intel GPU CI to README.md by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/562
* test split to 16, 32 by jp1924 in https://github.com/linkedin/Liger-Kernel/pull/564
* Clean up workaround introduced in PR 564 by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/566
* Update README.md by momochen in https://github.com/linkedin/Liger-Kernel/pull/567
* Grpo loss by kashif in https://github.com/linkedin/Liger-Kernel/pull/553
* Update Readme with ROCM installation instruction by zcnrex in https://github.com/linkedin/Liger-Kernel/pull/570
* fix qwen2vl and mllama test to pass failing tests by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/571
* KTO: Minor fix and documentation update by vaibhavjindal in https://github.com/linkedin/Liger-Kernel/pull/574
* Add TVD Loss Kernel by saurabhkoshatwar in https://github.com/linkedin/Liger-Kernel/pull/324
* Add KTO Benchmark Data into README by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/575
* Support Granite 3.0 and 3.1 models by JamesKunstle in https://github.com/linkedin/Liger-Kernel/pull/558
* Improve Hugging Face SFT Script by ParagEkbote in https://github.com/linkedin/Liger-Kernel/pull/539
* Add unit tests for shared prefix masked attention with `torch.FlexAttention` by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/504
* update project readme to include Granite support by JamesKunstle in https://github.com/linkedin/Liger-Kernel/pull/576
* Revert "Improve Hugging Face SFT Script (539)" and Fix TVD Test for Intel 580 by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/578
* Fix Rope Test by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/577
* Fix layer norm kernels by lancerts in https://github.com/linkedin/Liger-Kernel/pull/582
* Add OLMO2 model support by yundai424 in https://github.com/linkedin/Liger-Kernel/pull/581
* bump version to 0.5.4 by yundai424 in https://github.com/linkedin/Liger-Kernel/pull/585

New Contributors
* jp1924 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/564
* zcnrex made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/570
* vaibhavjindal made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/574
* saurabhkoshatwar made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/324
* JamesKunstle made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/558

**Full Changelog**: https://github.com/linkedin/Liger-Kernel/compare/v0.5.3...v0.5.4

What's Changed
* Add ref_input parameter to support separate inputs for reference model by xingyaoww in https://github.com/linkedin/Liger-Kernel/pull/467
* Revert "Add ref_input parameter to support separate inputs for reference model" by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/469
* Add dynamic dependency management for CUDA and ROCm by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/460
* [CI] runtime pip install using uv by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/471
* modify ref_input in chunked_loss base class and fix tests by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/470
* Add more post training in readme by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/472
* align post training loss at the center by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/473
* [Transformer] fix ORPO loss for MOE models by kashif in https://github.com/linkedin/Liger-Kernel/pull/479
* fix: correct typos in docstrings by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/482
* fix chosen_nll_loss in chunked losses by kashif in https://github.com/linkedin/Liger-Kernel/pull/486
* Revert "fix chosen_nll_loss in chunked losses (486)" by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/489
* fix dpo tests: reduce tolerance and change default compute_nll_loss false by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/490
* CPO & SimPO add label_smoothing by Mecoli1219 in https://github.com/linkedin/Liger-Kernel/pull/493
* Fix Preference Loss and Refactor for Readability by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/484
* annotate tl constexpr values by winglian in https://github.com/linkedin/Liger-Kernel/pull/497
* Fix Rope Compatibility with Cos/Sin Position Embedding for Batch Size > 1 by wizyoung in https://github.com/linkedin/Liger-Kernel/pull/477
* Move the checkstyle to [Ruff](https://docs.astral.sh/ruff/) by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/483
* Fix/liger fused linear cross entropy function does not support reduction=none by ryankert01 in https://github.com/linkedin/Liger-Kernel/pull/496
* Fix Dtype Mismatch in torch.addmm within ops/fused_linear_cross_entropy.py in AMP training. by DandinPower in https://github.com/linkedin/Liger-Kernel/pull/502
* Add weight support for LigerCrossEntropy by Tcc0403 in https://github.com/linkedin/Liger-Kernel/pull/420
* Refactor Temperature Scaling in Distillation Loss by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/444
* Fix All `chunked_loss` Benchmark Scripts by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/438
* Set z_loss_1d=None when return_z_loss=False in cross_entropy_loss to avoid tl.store fail when triton_interpret=1(for tl.device_print etc.) by wa008 in https://github.com/linkedin/Liger-Kernel/pull/508
* Add `aux_outputs` for CPO and SimPO by Mecoli1219 in https://github.com/linkedin/Liger-Kernel/pull/492
* Add `average_log_prob` args for cpo by Mecoli1219 in https://github.com/linkedin/Liger-Kernel/pull/510
* Refactor CrossEntropy and FusedLinearCrossEntropy by Tcc0403 in https://github.com/linkedin/Liger-Kernel/pull/511
* [ORPO] add nll_target for orpo nll loss by kashif in https://github.com/linkedin/Liger-Kernel/pull/503
* Format Benchmark Scripts with Ruff by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/516
* [Tiny] Add QVQ to readme by tyler-romero in https://github.com/linkedin/Liger-Kernel/pull/522
* Add argument `return_z_loss` to flce by Tcc0403 in https://github.com/linkedin/Liger-Kernel/pull/530
* Remove extra print by apaz-cli in https://github.com/linkedin/Liger-Kernel/pull/531
* Fix HF `transformers` Breaking Changes by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/526
* Handle cache_position for transformers 4.47.0 and later (528) by BenasdTW in https://github.com/linkedin/Liger-Kernel/pull/529
* Create Docs for Liger-Kernel by ParagEkbote in https://github.com/linkedin/Liger-Kernel/pull/485
* Add Mkdocs related dependencies to setup.py by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/534
* Add KTO Loss by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/475
* [tests] use a valid hexadecimal string instead of a placeholder by faaany in https://github.com/linkedin/Liger-Kernel/pull/535
* [tests] skip failed tests for xpu by faaany in https://github.com/linkedin/Liger-Kernel/pull/498
* Format files by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/541
* Fix Broken Links by ParagEkbote in https://github.com/linkedin/Liger-Kernel/pull/547
* [Fix] Fix the type hint of `test_utils::concatenated_forward` by hongpeng-guo in https://github.com/linkedin/Liger-Kernel/pull/549
* Add JSD Loss for Distillation by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/425
* [DPO] add reference log-prob outputs in DPO by kashif in https://github.com/linkedin/Liger-Kernel/pull/521
* Fix DPO unit test fail and refactor by Tcc0403 in https://github.com/linkedin/Liger-Kernel/pull/554

New Contributors
* xingyaoww made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/467
* kashif made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/479
* Mecoli1219 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/493
* winglian made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/497
* DandinPower made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/502
* wa008 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/508
* apaz-cli made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/531
* BenasdTW made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/529
* ParagEkbote made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/485

**Full Changelog**: https://github.com/linkedin/Liger-Kernel/compare/v0.5.2...v0.5.3

What's Changed
* Disable Qwen2 VL test for with logits conv test by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/463
* Fix Qwen2VL mrope for transformers 4.47.0 by li-plus in https://github.com/linkedin/Liger-Kernel/pull/464
* Revert Workaround of Disabling QWEN2_VL in Convergence Tests by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/466


**Full Changelog**: https://github.com/linkedin/Liger-Kernel/compare/v0.5.1...v0.5.2

What's Changed
* Fix liger orpo trainer import error by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/459
* Update pyproject.toml by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/462


**Full Changelog**: https://github.com/linkedin/Liger-Kernel/compare/v0.5.0...v0.5.1

Highlights
1. **Post Training Loss**: Introducing the first open-source optimized post-training losses in Liger Kernel with ~80% memory reduction, featuring DPO, CPO, ORPO, SimPO, JSD, and more. No more OOM nightmares for post-training ML researchers!
<img src="https://github.com/user-attachments/assets/19efbd07-f70b-4573-a3be-33fa80e7c4e1" alt="image" width="400">

2. **AMD CI**: With AMD’s generous sponsorship of MI300s, we’ve integrated them into our CI. Special thanks to Embedded LLM for building the AMD CI infrastructure. 428
3. **XPU Support**: In collaboration with Intel, we now support XPU, demonstrating comparable performance gains with other vendors. 407

What's Changed
* Adds the CPO Alignment Loss Function by pramodith in https://github.com/linkedin/Liger-Kernel/pull/382
* Qwen2-VL Training Example w/ Liger by tyler-romero in https://github.com/linkedin/Liger-Kernel/pull/389
* Support Qwen2-VL's multimodal RoPE implementation by li-plus in https://github.com/linkedin/Liger-Kernel/pull/384
* add xpu device support for `rms_norm` by faaany in https://github.com/linkedin/Liger-Kernel/pull/379
* fix qwen2 import failure in test by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/394
* Add Chunked SimPO Loss by pramodith in https://github.com/linkedin/Liger-Kernel/pull/386
* Add script to reproducibly run examples on Modal by tyler-romero in https://github.com/linkedin/Liger-Kernel/pull/397
* add nn.module support for chunked loss function by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/402
* Generalize JSD to FKL/RKL by yundai424 in https://github.com/linkedin/Liger-Kernel/pull/393
* Enable keyword arguments for liger functional by hongpeng-guo in https://github.com/linkedin/Liger-Kernel/pull/400
* add reference model logps to chunkedloss interface and fix dpo loss fn by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/405
* Optimize CE Loss by casting dtype to float32 inside kernel by pramodith in https://github.com/linkedin/Liger-Kernel/pull/406
* Xpu support by mgrabban in https://github.com/linkedin/Liger-Kernel/pull/407
* Fix `get_batch_loss_metrics` comments by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/413
* Add rebuild to CI by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/415
* Fix os env by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/416
* Adjust QWEN2 VL Loss `rtol` by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/412
* [tiny] Add QwQ to readme (same arch as Qwen2) by tyler-romero in https://github.com/linkedin/Liger-Kernel/pull/424
* Enhance Cross Entropy Softcap Unit Test by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/423
* Add ORPO Trainer + support HF metrics directly from chunked loss functions + fixes to avoid torch compile recompilations by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/429
* Add Build Success/Fail Badge by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/431
* Switch amd-ci to use MI300X runner. by saienduri in https://github.com/linkedin/Liger-Kernel/pull/428
* [CI] rename ci and add cron job for amd by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/433
* [CI] shorten ci name by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/434
* update ci icon on readme by bboyleonp666 in https://github.com/linkedin/Liger-Kernel/pull/440
* Introduce Knowledge Distillation Base by austin362667 in https://github.com/linkedin/Liger-Kernel/pull/432
* [AMD] [CI] Clean up `amd-ci` by tjtanaa in https://github.com/linkedin/Liger-Kernel/pull/436
* Add xpu in env report by abhilash1910 in https://github.com/linkedin/Liger-Kernel/pull/443
* Specify scheduled CI in AMD badge by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/446
* improve code quality for chunk loss by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/448
* Add paper link and formula for preference loss by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/449
* Make kernel doc lean by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/450
* Fix LigerCrossEntropyLoss Reduction Behavior for "None" Mode by hebiao064 in https://github.com/linkedin/Liger-Kernel/pull/435
* add eng blog by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/452
* add chunked loss to readme by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/453
* change chunked readme by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/454
* add sponsorship and collab by ByronHsu in https://github.com/linkedin/Liger-Kernel/pull/457
* version bump to 0.5.0 by shivam15s in https://github.com/linkedin/Liger-Kernel/pull/455
* Add HIP (ROCm) and Liger Kernel to env report by Comet0322 in https://github.com/linkedin/Liger-Kernel/pull/456

New Contributors
* li-plus made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/384
* faaany made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/379
* hongpeng-guo made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/400
* mgrabban made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/407
* hebiao064 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/431
* saienduri made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/428
* bboyleonp666 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/440
* abhilash1910 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/443
* Comet0322 made their first contribution in https://github.com/linkedin/Liger-Kernel/pull/456