liger-kernel 0.5.2__py3-none-any.whl → 0.5.4__py3-none-any.whl

liger_kernel/transformers/trainer/orpo_trainer.py

@@ -1,7 +1,14 @@
-from typing import Any, Callable, Dict, List, Literal, Tuple, Union
+from typing import Any
+from typing import Callable
+from typing import Dict
+from typing import List
+from typing import Literal
+from typing import Tuple
+from typing import Union
 
 import torch
 import torch.nn as nn
+
 from torch.distributed.fsdp import FullyShardedDataParallel
 from trl.trainer import ORPOTrainer
 
@@ -17,7 +24,7 @@ class _FSDPForwardRedirection:
     This is needed in cases where we call a submodule of a FSDP module. For instance, when we want to call only
     the `LlamaModel` part out of a FSDP-wrapped `LlamaForCausalLM` to get the hidden states without involving
     GPU-memory-heavy `lm_head` and cross entropy computation, doing this directly (i.e. `model.model.forward()`)
-    will not work because the first `nn.Emebedding` layer is not independently wrapped as a FSDP module (because of
+    will not work because the first `nn.Embedding` layer is not independently wrapped as a FSDP module (because of
     the transformer-based wrapping policy), and not calling it through FSDP root module forward will not all-gather
     its parameter, thus resulting in "RuntimeError: 'weight' must be 2-D" error. Similarly, if we want to call just
     the `lm_head` part of a model, we need this trick too to properly get its params all-gathered.
@@ -62,9 +69,7 @@ class _FSDPForwardRedirection:
 class LigerORPOTrainer(ORPOTrainer):
     def concatenated_forward(
         self, model: nn.Module, batch: Dict[str, Union[List, torch.LongTensor]]
-    ) -> Tuple[
-        torch.FloatTensor, torch.FloatTensor, torch.FloatTensor, torch.FloatTensor
-    ]:
+    ) -> Tuple[torch.FloatTensor, torch.FloatTensor, torch.FloatTensor, torch.FloatTensor]:
         """
         Run the given model on the given batch of inputs, concatenating the chosen and rejected inputs together.
         We do this to avoid doing two forward passes, because it's faster for FSDP.
@@ -79,9 +84,7 @@ class LigerORPOTrainer(ORPOTrainer):
 
         model_kwargs = (
             {
-                "decoder_input_ids": self._shift_right(
-                    concatenated_batch["concatenated_labels"]
-                ),
+                "decoder_input_ids": self._shift_right(concatenated_batch["concatenated_labels"]),
             }
             if self.is_encoder_decoder
             else {}
@@ -90,6 +93,13 @@ class LigerORPOTrainer(ORPOTrainer):
         if self.aux_loss_enabled:
             model_kwargs["output_router_logits"] = True
 
+        if self.is_encoder_decoder:
+            labels = concatenated_batch["concatenated_labels"].clone()
+        else:
+            labels = concatenated_batch["concatenated_input_ids"].clone()
+            attention_mask = concatenated_batch["concatenated_attention_mask"]
+            labels = torch.where(attention_mask == 1, labels, self.label_pad_token_id)
+
         if isinstance(model, FullyShardedDataParallel):
             outputs = _FSDPForwardRedirection()(
                 model,
@@ -109,22 +119,27 @@ class LigerORPOTrainer(ORPOTrainer):
                 **model_kwargs,
             )
 
-        orpo_loss_fn = LigerFusedLinearORPOLoss(
-            ignore_index=self.label_pad_token_id, beta=self.beta
-        )
+        orpo_loss_fn = LigerFusedLinearORPOLoss(ignore_index=self.label_pad_token_id, beta=self.beta)
 
-        def orpo_partial(lm_head, last_hidden_state, concatenated_labels):
+        def orpo_partial(lm_head, last_hidden_state, concatenated_labels, nll_target):
             return orpo_loss_fn(
-                lm_head.weight, last_hidden_state, concatenated_labels, lm_head.bias
+                lm_head.weight, last_hidden_state, concatenated_labels, lm_head.bias, nll_target=nll_target
             )
 
         orpo_loss, aux_outputs = _FSDPForwardRedirection()(
             model,
             orpo_partial,
             model.lm_head,
-            outputs.last_hidden_state,
-            concatenated_batch["concatenated_labels"],
+            outputs.last_hidden_state[:, :-1] if not self.is_encoder_decoder else outputs.last_hidden_state,
+            concatenated_batch["concatenated_labels"][:, 1:]
+            if not self.is_encoder_decoder
+            else concatenated_batch["concatenated_labels"],
+            labels[:, 1:] if not self.is_encoder_decoder else labels,
         )
+        # if aux_loss_enabled, add the aux_loss to the orpo_loss
+        if self.aux_loss_enabled:
+            orpo_loss += self.aux_loss_coef * outputs.aux_loss
+
         return orpo_loss, aux_outputs
 
     def get_batch_loss_metrics(
@@ -145,9 +160,7 @@ class LigerORPOTrainer(ORPOTrainer):
         ) = aux_outputs[:5]
 
         # return loss, metrics
-        chosen_rewards, rejected_rewards, log_odds_ratio, log_odds_chosen = aux_outputs[
-            5:
-        ]
+        chosen_rewards, rejected_rewards, log_odds_ratio, log_odds_chosen = aux_outputs[5:]
 
         reward_accuracies = (chosen_rewards > rejected_rewards).float()
 

liger_kernel/transformers/tvd.py

@@ -0,0 +1,13 @@
+import torch.nn as nn
+
+from liger_kernel.ops.tvd import LigerTVDLossFunction
+
+
+class LigerTVDLoss(nn.Module):
+    def __init__(self, reduction="batchmean", ignore_index: int = -100):
+        super(LigerTVDLoss, self).__init__()
+        self.reduction = reduction
+        self.ignore_index = ignore_index
+
+    def forward(self, p, q, shift_labels=None):
+        return LigerTVDLossFunction.apply(p, q, shift_labels, self.reduction, self.ignore_index)

liger_kernel/triton/__init__.py

@@ -1,3 +1 @@
-from liger_kernel.triton.monkey_patch import (  # noqa: F401
-    apply_liger_triton_cache_manager,
-)
+from liger_kernel.triton.monkey_patch import apply_liger_triton_cache_manager  # noqa: F401

liger_kernel/triton/monkey_patch.py

@@ -37,6 +37,4 @@ def apply_liger_triton_cache_manager():
     Experimental feature to get around transient FileNotFoundError in triton compilation.
     For more details please see https://github.com/triton-lang/triton/pull/4295
     """
-    os.environ["TRITON_CACHE_MANAGER"] = (
-        "liger_kernel.triton.monkey_patch:LigerTritonFileCacheManager"
-    )
+    os.environ["TRITON_CACHE_MANAGER"] = "liger_kernel.triton.monkey_patch:LigerTritonFileCacheManager"

liger_kernel/utils.py

@@ -9,5 +9,54 @@ def infer_device():
         return "cuda"
     elif torch.xpu.is_available():
         return "xpu"
+    elif torch.hip.is_available():
+        return "hip"
     else:
         return "cpu"
+
+
+def transformers_version_dispatch(
+    required_version: str,
+    before_fn,
+    after_fn,
+    before_args: tuple = (),
+    after_args: tuple = (),
+    before_kwargs: dict = None,
+    after_kwargs: dict = None,
+):
+    """
+    Dispatches to different functions based on package version comparison.
+
+    Args:
+        required_version: Version to compare against (e.g. "4.48.0")
+        before_fn: Function to call if package_version < required_version
+        after_fn: Function to call if package_version >= required_version
+        before_args: Positional arguments for before_fn
+        after_args: Positional arguments for after_fn
+        before_kwargs: Keyword arguments for before_fn
+        after_kwargs: Keyword arguments for after_fn
+
+    Returns:
+        Result from either before_fn or after_fn
+
+    Example:
+        >>> rotary_emb = transformers_version_dispatch(
+        ...     "4.48.0",
+        ...     LlamaRotaryEmbedding,
+        ...     LlamaRotaryEmbedding,
+        ...     before_args=(head_dim,),
+        ...     after_args=(LlamaConfig(head_dim=head_dim),),
+        ...     before_kwargs={'device': device},
+        ...     after_kwargs={'device': device}
+        ... )
+    """
+    from packaging import version
+    from transformers import __version__ as transformers_version
+
+    before_kwargs = before_kwargs or {}
+    after_kwargs = after_kwargs or {}
+
+    if version.parse(transformers_version) < version.parse(required_version):
+        return before_fn(*before_args, **before_kwargs)
+    else:
+        return after_fn(*after_args, **after_kwargs)

{liger_kernel-0.5.2.dist-info → liger_kernel-0.5.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: liger_kernel
-Version: 0.5.2
+Version: 0.5.4
 Summary: Efficient Triton kernels for LLM Training
 License: BSD 2-CLAUSE LICENSE
         Copyright 2024 LinkedIn Corporation 
@@ -32,10 +32,6 @@ License-File: LICENSE
 License-File: NOTICE
 Requires-Dist: torch>=2.1.2
 Requires-Dist: triton>=2.3.1
-Provides-Extra: transformers
-Requires-Dist: transformers~=4.0; extra == "transformers"
-Provides-Extra: trl
-Requires-Dist: trl>=0.11.0; extra == "trl"
 Provides-Extra: dev
 Requires-Dist: transformers>=4.44.2; extra == "dev"
 Requires-Dist: matplotlib>=3.7.2; extra == "dev"
@@ -46,13 +42,11 @@ Requires-Dist: pytest>=7.1.2; extra == "dev"
 Requires-Dist: pytest-xdist; extra == "dev"
 Requires-Dist: pytest-rerunfailures; extra == "dev"
 Requires-Dist: datasets>=2.19.2; extra == "dev"
-Requires-Dist: torchvision>=0.16.2; extra == "dev"
 Requires-Dist: seaborn; extra == "dev"
-Provides-Extra: amd
-Requires-Dist: torch>=2.6.0.dev; extra == "amd"
-Requires-Dist: setuptools-scm>=8; extra == "amd"
-Requires-Dist: torchvision>=0.20.0.dev; extra == "amd"
-Requires-Dist: triton>=3.0.0; extra == "amd"
+Requires-Dist: mkdocs; extra == "dev"
+Requires-Dist: mkdocs-material; extra == "dev"
+Dynamic: provides-extra
+Dynamic: requires-dist
 
 <a name="readme-top"></a>
 
@@ -103,6 +97,11 @@ Requires-Dist: triton>=3.0.0; extra == "amd"
                     <img src="https://github.com/linkedin/Liger-Kernel/actions/workflows/amd-ci.yml/badge.svg?event=schedule" alt="Build">
                 </a>
             </div>
+            <div style="display: block;">
+                <a href="https://github.com/linkedin/Liger-Kernel/actions/workflows/amd-ci.yml">
+                    <img src="https://github.com/linkedin/Liger-Kernel/actions/workflows/intel-ci.yml/badge.svg?event=schedule" alt="Build">
+                </a>
+            </div>
         </td>
     </tr>
 </table>
@@ -116,7 +115,8 @@ Requires-Dist: triton>=3.0.0; extra == "amd"
 <details>
   <summary>Latest News 🔥</summary>
 
-  - [2024/12/15] We release LinkedIn Engineering Blog - [Liger-Kernel: Empowering an open source ecosystem of Triton Kernels for Efficient LLM Training](https://www.linkedin.com/blog/engineering/open-source/liger-kernel-open-source-ecosystem-for-efficient-llm-training)
+  - [2024/12/11] We release [v0.5.0](https://github.com/linkedin/Liger-Kernel/releases/tag/v0.5.0): 80% more memory efficient post training losses (DPO, ORPO, CPO, etc)!
+  - [2024/12/5] We release LinkedIn Engineering Blog - [Liger-Kernel: Empowering an open source ecosystem of Triton Kernels for Efficient LLM Training](https://www.linkedin.com/blog/engineering/open-source/liger-kernel-open-source-ecosystem-for-efficient-llm-training)
   - [2024/11/6] We release [v0.4.0](https://github.com/linkedin/Liger-Kernel/releases/tag/v0.4.0): Full AMD support, Tech Report, Modal CI, Llama-3.2-Vision!
   - [2024/10/21] We have released the tech report of Liger Kernel on Arxiv: https://arxiv.org/pdf/2410.10989
   - [2024/9/6] We release v0.2.1 ([X post](https://x.com/liger_kernel/status/1832168197002510649)). 2500+ Stars, 10+ New Contributors, 50+ PRs, 50k Downloads in two weeks!
@@ -128,7 +128,7 @@ Requires-Dist: triton>=3.0.0; extra == "amd"
 
 **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.
 
-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. 
+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).
 
 ## Supercharge Your Model with Liger Kernel
 
@@ -145,6 +145,21 @@ With one line of code, Liger Kernel can increase throughput by more than 20% and
 > - Benchmark conditions: LLaMA 3-8B, Batch Size = 8, Data Type = `bf16`, Optimizer = AdamW, Gradient Checkpointing = True, Distributed Strategy = FSDP1 on 8 A100s.
 > - Hugging Face models start to OOM at a 4K context length, whereas Hugging Face + Liger Kernel scales up to 16K.
 
+## Optimize Post Training with Liger Kernel
+
+<p align="center">
+    <img src="https://raw.githubusercontent.com/linkedin/Liger-Kernel/main/docs/images/post-training.png" width="50%" alt="Post Training">
+</p>
+
+We provide optimized post training kernels like DPO, ORPO, SimPO, and more which can reduce memory usage by up to 80%. You can easily use them as python modules.
+
+```python
+from liger_kernel.chunked_loss import LigerFusedLinearDPOLoss
+orpo_loss = LigerFusedLinearORPOLoss()
+y = orpo_loss(lm_head.weight, x, target)
+```
+
+
 ## Examples
 
 | **Use Case**                                    | **Description**                                                                                   |
@@ -178,6 +193,11 @@ With one line of code, Liger Kernel can increase throughput by more than 20% and
 - `torch >= 2.5.0` Install according to the instruction in Pytorch official webpage.
 - `triton >= 3.0.0` Install from pypi. (e.g. `pip install triton==3.0.0`)
 
+```bash
+# Need to pass the url when installing
+pip install -e .[dev] --extra-index-url https://download.pytorch.org/whl/nightly/rocm6.2
+```
+
 ### Optional Dependencies
 
 - `transformers >= 4.x`: Required if you plan to use the transformers models patching APIs. The specific model you are working will dictate the minimum version of transformers.
@@ -202,11 +222,13 @@ To install from source:
 ```bash
 git clone https://github.com/linkedin/Liger-Kernel.git
 cd Liger-Kernel
+
+# Install Default Dependencies
+# Setup.py will detect whether you are using AMD or NVIDIA
 pip install -e .
-# or if installing on amd platform
-pip install -e .[amd] --extra-index-url https://download.pytorch.org/whl/nightly/rocm6.2 # rocm6.2
-# or if using transformers
-pip install -e .[transformers]
+
+# Setup Development Dependencies
+pip install -e ".[dev]"
 ```
 
 
@@ -252,7 +274,7 @@ model = transformers.AutoModelForCausalLM("path/to/llama/model")
 
 ### 3. Compose Your Own Model
 
-You can take individual [kernels](#kernels) to compose your models.
+You can take individual [kernels](https://github.com/linkedin/Liger-Kernel?tab=readme-ov-file#model-kernels) to compose your models.
 
 ```python
 from liger_kernel.transformers import LigerFusedLinearCrossEntropyLoss
@@ -291,8 +313,10 @@ loss.backward()
 | Gemma1      | `liger_kernel.transformers.apply_liger_kernel_to_gemma`    | RoPE, RMSNorm, GeGLU, CrossEntropyLoss, FusedLinearCrossEntropy         |
 | Gemma2      | `liger_kernel.transformers.apply_liger_kernel_to_gemma2`   | RoPE, RMSNorm, GeGLU, CrossEntropyLoss, FusedLinearCrossEntropy         |
 | Qwen2, Qwen2.5, & QwQ      | `liger_kernel.transformers.apply_liger_kernel_to_qwen2`    | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy        |
-| Qwen2-VL       | `liger_kernel.transformers.apply_liger_kernel_to_qwen2_vl`    | RMSNorm, LayerNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy        |
+| Qwen2-VL, & QVQ       | `liger_kernel.transformers.apply_liger_kernel_to_qwen2_vl`    | RMSNorm, LayerNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy        |
 | Phi3 & Phi3.5       | `liger_kernel.transformers.apply_liger_kernel_to_phi3`     | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy         |
+| Granite 3.0 & 3.1   | `liger_kernel.transformers.apply_liger_kernel_to_granite`     | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss |
+| OLMo2   | `liger_kernel.transformers.apply_liger_kernel_to_olmo2`     | RoPE, RMSNorm, SwiGLU, CrossEntropyLoss, FusedLinearCrossEntropy |
 
 
 ## Low-level APIs
@@ -321,6 +345,7 @@ loss.backward()
 | Fused Linear DPO Loss           | `liger_kernel.chunked_loss.LigerFusedLinearDPOLoss`       |
 | Fused Linear ORPO Loss          | `liger_kernel.chunked_loss.LigerFusedLinearORPOLoss`      |
 | Fused Linear SimPO Loss         | `liger_kernel.chunked_loss.LigerFusedLinearSimPOLoss`     |
+| Fused Linear KTO Loss           | `liger_kernel.chunked_loss.LigerFusedLinearKTOLoss`     |
 
 ### Distillation Kernels
 
@@ -329,6 +354,7 @@ loss.backward()
 | KLDivergence                    | `liger_kernel.transformers.LigerKLDIVLoss`                  |
 | JSD                             | `liger_kernel.transformers.LigerJSD`                        |
 | Fused Linear JSD                  | `liger_kernel.transformers.LigerFusedLinearJSD`             |
+| TVD                             | `liger_kernel.transformers.LigerTVDLoss`                    |
 
 ### Experimental Kernels
 
@@ -340,16 +366,17 @@ loss.backward()
 
 ## Contributing, Acknowledgements, and License
 
-- [Contributing Guidelines](https://github.com/linkedin/Liger-Kernel/blob/main/docs/CONTRIBUTING.md)
-- [Acknowledgements](https://github.com/linkedin/Liger-Kernel/blob/main/docs/Acknowledgement.md)
-- [License Information](https://github.com/linkedin/Liger-Kernel/blob/main/docs/License.md)
+- [Contributing Guidelines](https://github.com/linkedin/Liger-Kernel/blob/main/docs/contributing.md)
+- [Acknowledgements](https://github.com/linkedin/Liger-Kernel/blob/main/docs/acknowledgement.md)
+- [License Information](https://github.com/linkedin/Liger-Kernel/blob/main/docs/license.md)
 
 ## Sponsorship and Collaboration
- 
+
+- [Glows.ai](https://platform.glows.ai/): Sponsoring NVIDIA GPUs for our open source developers.
 - [AMD](https://www.amd.com/en.html): Providing AMD GPUs for our AMD CI.
 - [Intel](https://www.intel.com/): Providing Intel GPUs for our Intel CI.
 - [Modal](https://modal.com/): Free 3000 credits from GPU MODE IRL for our NVIDIA CI.
-- [EmbeddedLLM](https://embeddedllm.com/): Making Liger Kernel run fast and stable on AMD. 
+- [EmbeddedLLM](https://embeddedllm.com/): Making Liger Kernel run fast and stable on AMD.
 - [HuggingFace](https://huggingface.co/): Integrating Liger Kernel into Hugging Face Transformers and TRL.
 - [Lightning AI](https://lightning.ai/): Integrating Liger Kernel into Lightning Thunder.
 - [Axolotl](https://axolotl.ai/): Integrating Liger Kernel into Axolotl.
@@ -359,7 +386,7 @@ loss.backward()
 
 - For issues, create a Github ticket in this repository
 - For open discussion, join [our discord channel](https://discord.gg/gpumode)
-- For formal collaboration, send an email to byhsu@linkedin.com
+- For formal collaboration, send an email to yannchen@linkedin.com
 
 ## Cite this work
 

liger_kernel-0.5.4.dist-info/RECORD

@@ -0,0 +1,74 @@
+liger_kernel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+liger_kernel/env_report.py,sha256=uhdEC8OydxoZlb7B6YYcAaBF3crGFdIck-4cxaW4NJY,1728
+liger_kernel/utils.py,sha256=FtVUkCGBT1UNasTl6HMNycWwiwHayK6tx-ZDdA-sNX4,1884
+liger_kernel/chunked_loss/README.md,sha256=0FmkFC3hKBqyoDT5uTlIYmrvRkF-EOCR1y-EBU1LpWU,2248
+liger_kernel/chunked_loss/__init__.py,sha256=ATu-xX5Fc49Cr6yBOGBRNTo593ZrU5ZCsIuvoIbJWw4,603
+liger_kernel/chunked_loss/cpo_loss.py,sha256=OdBR8WYdHTKpLI_c9DcuwqKSWPeAAeTyREz46Vu_cAY,3682
+liger_kernel/chunked_loss/dpo_loss.py,sha256=wgjnwzLfrMUwV5mXgrq6G1YfQKWnbiFJegaP48BGJHY,4509
+liger_kernel/chunked_loss/functional.py,sha256=THWWpCnRVhTVfnPnyvQjdBvo1JDtxhwLmtZE_yiBBqM,817
+liger_kernel/chunked_loss/fused_linear_distillation.py,sha256=5V8rdva89WyHVbmJ8JOmC4DYNOR6ByXfx3qlUieOZkI,11002
+liger_kernel/chunked_loss/fused_linear_preference.py,sha256=idK9V9NivoVITqVpiG0fEGUHSvinYWkn9-EYXZjR-KQ,18356
+liger_kernel/chunked_loss/fused_linear_rlhf.py,sha256=sAApL4GQ3YL2F-ymIAF61GCpFfBgFcWF5LB4Gzd7LgY,8044
+liger_kernel/chunked_loss/fused_linear_unpaired_preference.py,sha256=ZqYlXXhIphkJPxOS7iI70avgrr6x0skEtgpckZTYau0,9819
+liger_kernel/chunked_loss/grpo_loss.py,sha256=M5qlQR-v5Rh8N3P3dPGNhOKygDFJ4516_rJaVPzU_-c,4980
+liger_kernel/chunked_loss/jsd_loss.py,sha256=yRCQdvd3ruTWP4A_BfU8VcZ6LepSUfO0Ob7stGnueQY,6052
+liger_kernel/chunked_loss/kto_loss.py,sha256=b3ffJyk97e-6XdXd4HFrYyx8wW4A-CU4gOaJSimKLtA,5476
+liger_kernel/chunked_loss/orpo_loss.py,sha256=yjcrrbVeemLYodoSKT-FMSnaPtyKAZ3aOrvPD6tTY6Y,3617
+liger_kernel/chunked_loss/simpo_loss.py,sha256=3TTc7U79Orjgi-Wu81WZkWk5MgsdqKXIOBHgIvDazPw,3865
+liger_kernel/ops/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+liger_kernel/ops/cross_entropy.py,sha256=D6vFFloiuxFXoWfjlIjmfO3tVaWOiYmztw9FKAi5vdU,18608
+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/geglu.py,sha256=axGvCIvlBzuluoAIrWTsp2iZM4BFKNInkPov8YVvH9E,4126
+liger_kernel/ops/group_norm.py,sha256=qD4D4lSjSgVtO52EBNLC2iTseALRgPgqXE50U2woggk,10837
+liger_kernel/ops/jsd.py,sha256=0jNeRxpcNI5ckxCdoCNyO5GEedLIuzx3lz6KAiksc4o,6109
+liger_kernel/ops/kl_div.py,sha256=MnfuYqqQESON1X2Swy064x1urKtMFdgeSWd60VttBXI,8420
+liger_kernel/ops/layer_norm.py,sha256=6roQjioyg-9O2qLPV8nL4U0-5UH80tdzOMTWwjvDnn8,7961
+liger_kernel/ops/qwen2vl_mrope.py,sha256=3GExhYpLgB4VUtyZyjRk8XjEur3W4EWF6HQ67ML5vBU,8481
+liger_kernel/ops/rms_norm.py,sha256=PWLJcdIKU5e-8BuYFHd9Cqlq6wmr6fUXKi9zQD4LetU,11727
+liger_kernel/ops/rope.py,sha256=ofmBOkUpZZO-Q8Z5B_LOFYYLD-YT-8WnJ4vGOrDYouI,8943
+liger_kernel/ops/swiglu.py,sha256=KmgMjaJQnbLLgZn2nEpbwHU_xpnYRweCyrLQSVvM1vA,3015
+liger_kernel/ops/tvd.py,sha256=FHJtLQI95ijqgg9UtaHpMAjSCiPxB6CduPwPMcGxelc,6405
+liger_kernel/ops/utils.py,sha256=uoFKQqo-34N2TWQNvXMFywqGiOMMXNEVBxVojzlUAa0,3836
+liger_kernel/ops/experimental/embedding.py,sha256=tolj3tItkzpSb30zWqDN2_yX4ectflaQ8HMyKyFIQc8,4172
+liger_kernel/ops/experimental/mm_int8int2.py,sha256=TrS9lpwekrik_w5qE7AhMJD1bcq-OidjtbsW80oZ6IM,13314
+liger_kernel/transformers/__init__.py,sha256=6v_VcV1GQ9ISgNCd-ZxtmEg_s5GTBQ9F-s1KrFkYzPQ,2265
+liger_kernel/transformers/auto_model.py,sha256=0qCTRZt280Bj_LcFdzo9hlaR-BWNazawXOGgoCZjgEg,1545
+liger_kernel/transformers/cross_entropy.py,sha256=z3KTWQnFxr_IZaVjtYt0ZNEWQdDdYThN35xWkHlDGH0,1683
+liger_kernel/transformers/functional.py,sha256=ShLD3eb--XKNtllznCrOYTbo4f-1KVwzi0KLMICdrn4,4942
+liger_kernel/transformers/fused_linear_cross_entropy.py,sha256=09Rt7FZzLH42VOcIbQ4dlQd0o3Rlb4vk6fqiOQ7WTD8,1778
+liger_kernel/transformers/fused_linear_jsd.py,sha256=bZ4otCvWBuOnA5XdQL-FzZVItJlDt-ht9e_pG7PG93E,3999
+liger_kernel/transformers/geglu.py,sha256=mrgqzIUVd6lN7fkDKLkw5YaESDxDtFgbot430WwPVOQ,1107
+liger_kernel/transformers/group_norm.py,sha256=6qMAWOprr4SzP0YhNVNGQIBpM5aUHplUD2VuGJrMBz0,2173
+liger_kernel/transformers/jsd.py,sha256=DGqRnxIZxsvxo0_tbbxX3b-sDbDjC_yKufyRIHCcScY,2979
+liger_kernel/transformers/kl_div.py,sha256=WLffFbh1EExD2Eb1F7lN11fo9JJC-0751WJjZAF1Fj8,409
+liger_kernel/transformers/layer_norm.py,sha256=c9pk3PEasOKYR0rhe5e5nNrnYKVCEW4VC8S6LpCq9EQ,906
+liger_kernel/transformers/monkey_patch.py,sha256=g3i3q5McBg23A3Mnviw-Eb32le1hvN7jByzONa9ngcs,44000
+liger_kernel/transformers/qwen2vl_mrope.py,sha256=5EwSqrMdsL9MYspeBMXBsNJKvH0MOmRrtJXAJlnnlOI,1047
+liger_kernel/transformers/rms_norm.py,sha256=GqCEJuGt0YdqqlMcToE0Wp4A8YFquDa4UUSyH2uFW2A,1191
+liger_kernel/transformers/rope.py,sha256=ZTrTORSAyfcFIKjk6XEeYmk4ROH7xXED9L4g2NFntlE,999
+liger_kernel/transformers/swiglu.py,sha256=i9WTqcNRqReU4XJs391IPbl-I5X0wG4T72D4pqGFfJg,2422
+liger_kernel/transformers/trainer_integration.py,sha256=W3ON51O5GkyzNJsItz0y5rKx-uy2f2cFfveZpqbUdhw,123
+liger_kernel/transformers/tvd.py,sha256=XrRfyJIqN6HFxXk8MYyFVZM1OLz3mtSbRZvWfZ_JerQ,450
+liger_kernel/transformers/experimental/embedding.py,sha256=2P0QYdlFyFrG5OqTzTa1wcRgDSyjBMv5i1a7BrDPDQw,881
+liger_kernel/transformers/model/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+liger_kernel/transformers/model/gemma.py,sha256=ky89b3aWPaeTGRMC-745KgixtQIRXzNAiCORAMLn9yo,9654
+liger_kernel/transformers/model/gemma2.py,sha256=27NcoZjEqP9Lqb4Wf0EKqTbr2HTGiHPhrVyPCRGPz6s,10767
+liger_kernel/transformers/model/llama.py,sha256=3LJFXKFDKvEakaWPc_NicSFst4Y_hdSMrdl1UDK1EcA,10330
+liger_kernel/transformers/model/mistral.py,sha256=MVRksI5_j_8WJu8znOHKCdSI5jSu-S7cdFYzt9m_vIQ,5180
+liger_kernel/transformers/model/mixtral.py,sha256=jpZJkpl625Q-JHWarj2MqT5mRaSsiCtg0c9vVyvOdCY,11430
+liger_kernel/transformers/model/mllama.py,sha256=qWexBdskuN3gPJvPUwt4J0nU675tGD6W7wxgRZ9Bifg,11145
+liger_kernel/transformers/model/olmo2.py,sha256=yyksS6E4fuWd8asEW8rEDBKqZpFmP4ITCM_bjIDZaoY,5124
+liger_kernel/transformers/model/phi3.py,sha256=biRa8fph9qdnQmkD9I21t5XIjpIt1i6UKU4uk8Up8pU,10292
+liger_kernel/transformers/model/qwen2.py,sha256=14UuPjxB-tjqWn85Tn4fqBFvVhVsth5iPEt8kJSMiew,9581
+liger_kernel/transformers/model/qwen2_vl.py,sha256=yMLqsfSYcvhClUpTUjGoADiOxfLB2B8240VdrPP0c8s,9851
+liger_kernel/transformers/trainer/__init__.py,sha256=p7yQfklV8-467qSz_ZMimkbDF7HHWHwku25A-GYL0WU,193
+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-0.5.4.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
+liger_kernel-0.5.4.dist-info/METADATA,sha256=Zw7n3Ey6vUed4E54H9-TzKmhuOpd9P2ZFMVL-zYUnew,22255
+liger_kernel-0.5.4.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
+liger_kernel-0.5.4.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+liger_kernel-0.5.4.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
+liger_kernel-0.5.4.dist-info/RECORD,,

{liger_kernel-0.5.2.dist-info → liger_kernel-0.5.4.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.6.0)
+Generator: setuptools (75.8.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

liger_kernel-0.5.2.dist-info/RECORD

@@ -1,65 +0,0 @@
-liger_kernel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-liger_kernel/env_report.py,sha256=FViyPju795lB6z4k2TZldvBSmQdcS0A2hcnDxepJrDo,1822
-liger_kernel/utils.py,sha256=HJa-xVKOohDn6pLVIx-Fv0V9h0QAL3qZGQNRICI-OpI,249
-liger_kernel/chunked_loss/__init__.py,sha256=R2wCcz4Y0kTAve926DH3k182XKezpXeACMHj05g9Mm8,346
-liger_kernel/chunked_loss/cpo_loss.py,sha256=Qu1Ul2A12sp6CqIT-atPbHWFb_LLtINEA9mOpIRx_0g,3097
-liger_kernel/chunked_loss/dpo_loss.py,sha256=H9_RRhclckHYM2sd75tgbnf8IxC_PU2JCALbgtPQvwc,4222
-liger_kernel/chunked_loss/functional.py,sha256=9Gr-YXIuEzEJkBUhDx3G2fuQayckLor7cC7svhmPML4,549
-liger_kernel/chunked_loss/fused_linear_distillation.py,sha256=2BH6DCPjsR2zS6zcwFPcIIZRhLF8SohjGdKsAJ_301o,10222
-liger_kernel/chunked_loss/fused_linear_preference.py,sha256=vlWfaaIECWvCQhY9PM7zRI0vKThIrydMf6P44bXn1EE,15114
-liger_kernel/chunked_loss/orpo_loss.py,sha256=ZuKGjbkIYzV4UzvupNdq6vyxCp7-BztQkUt8ZnFvKos,3531
-liger_kernel/chunked_loss/simpo_loss.py,sha256=Wa4LOlDG9PbJkOOkKg8hbKvnKgg7OTBz6-qIkwPK1yw,3275
-liger_kernel/ops/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-liger_kernel/ops/cross_entropy.py,sha256=oG5hfrlmnlF5lOoZRhHRglObxgH4B0KadjWMJj9EWPM,15860
-liger_kernel/ops/fused_linear_cross_entropy.py,sha256=Tnw4gyAYVVdnCOqhOuLEzbUQ3goOTnoAfk3pqSIM5ac,9301
-liger_kernel/ops/fused_linear_jsd.py,sha256=nOv4zwfxHqqepKEmMsQuz-B3H-gRjyo8uClpmqSGLYA,9693
-liger_kernel/ops/geglu.py,sha256=MQL4zyzneZqZYUGPvb1QjI_EYT9_pKfSDgR25WD9jrI,4127
-liger_kernel/ops/group_norm.py,sha256=VaRErVJGR4JqgXXvuIjNGTn3E2egjLtU1y3ymwIf4d8,10961
-liger_kernel/ops/jsd.py,sha256=Ap2b0_geCl6fqBXLI1IS6Yn6GlO-8LgPmnOW3y47dus,6151
-liger_kernel/ops/kl_div.py,sha256=03FNXfvCb6M-56hhFepAFV9p6brArPR6KOKkdGD34mw,8374
-liger_kernel/ops/layer_norm.py,sha256=_CZggw3GNEIUx5weDzadFit5I-Lzosoo8prgeJzcViY,7589
-liger_kernel/ops/qwen2vl_mrope.py,sha256=GvP4Cg-2ClYyiqbe7bB_OMvnlZooBmqP2-9V8RMPde4,8598
-liger_kernel/ops/rms_norm.py,sha256=g7OXwuYI8-LXudDwvXuiupVjjOsbu8c4wwv83VaHa54,11750
-liger_kernel/ops/rope.py,sha256=jrzaA9-6Orn44y_IIam9_YNPQxOFK2FrIRNfFea4EtU,8513
-liger_kernel/ops/swiglu.py,sha256=Fwxtd76rhHKT9ShQAGca9RsnASplAVxtYKHmiT73_yA,2994
-liger_kernel/ops/utils.py,sha256=_VQvd1PX5JXm5xaiBrk2gANp3qr4kM7qYG3ypkBwkMs,3850
-liger_kernel/ops/experimental/embedding.py,sha256=LYR66dB-jhvhtUjeV4PnNro-n77J1mdlmpSLSxB3Y6U,4186
-liger_kernel/ops/experimental/mm_int8int2.py,sha256=JpGVZCgRC6T8XMUJ_QbZRS2XU1bh0urIZphs5DTc1mY,13358
-liger_kernel/transformers/__init__.py,sha256=gia-eBxr7TLxU0GdDf8AfCY4WgDlFLqIGSt7EoQGsBA,1336
-liger_kernel/transformers/auto_model.py,sha256=RMIwQHSiXoksXFTIqFZ4PLBgoqkxJJAT3q1Qh47bGN8,1552
-liger_kernel/transformers/cross_entropy.py,sha256=yEm_YQ7oa3_BzT3hdW6KrAslduhSqWcJQVNZZDcWCg4,1758
-liger_kernel/transformers/functional.py,sha256=sUBoU8Vb4pLpr9G6IdkRsToYgh-rCXL4OLYat7Tv_GU,4450
-liger_kernel/transformers/fused_linear_cross_entropy.py,sha256=_i0PXSp5iZ9pKXdEeZ4lvHCENJYjV4y74yz3ZRG5XQg,1484
-liger_kernel/transformers/fused_linear_jsd.py,sha256=bZ4otCvWBuOnA5XdQL-FzZVItJlDt-ht9e_pG7PG93E,3999
-liger_kernel/transformers/geglu.py,sha256=QcrME_8ooIn0xa59LaC0aoOdRrBIFd11Y0bAyF0NfCw,1130
-liger_kernel/transformers/group_norm.py,sha256=FJ9R7mS9G1wO-GRIQ6QKSmIhnZ6nQ6GIkE4NnX_hnn0,2241
-liger_kernel/transformers/jsd.py,sha256=sbr8DnKSYZJH9pv2rpmboNijYGpZKbhb2-WSGp5_v6g,3001
-liger_kernel/transformers/kl_div.py,sha256=qVhjBg6tjRyue5iZ3NFxo8uySY4JuIFJyv0IM_50F24,431
-liger_kernel/transformers/layer_norm.py,sha256=fd6o4kSHJWolQMWxh-l1qObfgL08ruNbUoBiANKX1ow,972
-liger_kernel/transformers/monkey_patch.py,sha256=Fk2v4GZQDJzfh3Cpc6BHNJbs_tungDyWmqS9nuG9Lc4,38406
-liger_kernel/transformers/qwen2vl_mrope.py,sha256=5EwSqrMdsL9MYspeBMXBsNJKvH0MOmRrtJXAJlnnlOI,1047
-liger_kernel/transformers/rms_norm.py,sha256=AHstklNIO1PLHjjCBU-TPuUD-Fl_pycJUTLlJNojbV8,1189
-liger_kernel/transformers/rope.py,sha256=m-ah8vZBYW8tfplTXCiAPMHJWlB1tdp_JPXJeWE-Boo,943
-liger_kernel/transformers/swiglu.py,sha256=0-tVJ8xEYfhxnduc16PflXFj8sZPxdx9sHUn3hfwCI4,2468
-liger_kernel/transformers/trainer_integration.py,sha256=W3ON51O5GkyzNJsItz0y5rKx-uy2f2cFfveZpqbUdhw,123
-liger_kernel/transformers/experimental/embedding.py,sha256=HpckiAMKM8-SRxKDcGTqortVxnjhwpZsfsp9lfjqfeM,895
-liger_kernel/transformers/model/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-liger_kernel/transformers/model/gemma.py,sha256=R4huxuR48gkLrdT8KqV7As2v9dZtEmcGVz6YG1ZmuJE,9692
-liger_kernel/transformers/model/gemma2.py,sha256=zxQsxCRqkoxCES3GJPVI7soUuF3J5HZDlvJgaBos1zM,10836
-liger_kernel/transformers/model/llama.py,sha256=RinsgC_eR-YNvZd2SHPQxZ4eyR3uViaTFCM3SvI5nks,10426
-liger_kernel/transformers/model/mistral.py,sha256=XpL1rlWg_llvW3z_Hf_d8WQs7uQaH4ds7EZ2SxjQHsU,5144
-liger_kernel/transformers/model/mixtral.py,sha256=nyDS1dBpsOXYC2DuW59Hgu7ZrGftrHuWPfNqjcNPIxs,11503
-liger_kernel/transformers/model/mllama.py,sha256=mesNCgj0Ea1O-fqRD4LVxDJ1CR2abY_zAzK_bfVzkiU,11222
-liger_kernel/transformers/model/phi3.py,sha256=xUZPlaPKwknLjHc3uUW3EPodm1h0vD3G7Qnhh51v-Io,10332
-liger_kernel/transformers/model/qwen2.py,sha256=EyhSSzQOskGjSnCsKMZpd1s5IAIlHd5PBO3q0MoCs00,9619
-liger_kernel/transformers/model/qwen2_vl.py,sha256=bIQe2bWiY--G84FhCD29Gdi64_qHP6vbcGsK6vKysQE,8547
-liger_kernel/transformers/trainer/__init__.py,sha256=c4OQVJmhNOloj0JYSEc0j_cQuBbzGWILfaowUR1hmRw,210
-liger_kernel/transformers/trainer/orpo_trainer.py,sha256=jko6oq_XQdBSmXubp05E-_YXOyhtB5Bj75dg5YNwOsE,7517
-liger_kernel/triton/__init__.py,sha256=yfRe0zMb47QnqjecZWG7LnanfCTzeku7SgWRAwNVmzU,101
-liger_kernel/triton/monkey_patch.py,sha256=5BcGKTtdqeYchypBIBopGIWPx1-cFALz7sOKoEsqXJ0,1584
-liger_kernel-0.5.2.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
-liger_kernel-0.5.2.dist-info/METADATA,sha256=olSIT-Jd2Mowu2ja4QLwyPYBhCnY22znBq9pV7stkKI,20695
-liger_kernel-0.5.2.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
-liger_kernel-0.5.2.dist-info/WHEEL,sha256=PZUExdf71Ui_so67QXpySuHtCi3-J3wvF4ORK6k_S8U,91
-liger_kernel-0.5.2.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
-liger_kernel-0.5.2.dist-info/RECORD,,