liger-kernel-nightly 0.5.2.dev20241211055800__py3-none-any.whl → 0.5.2.dev20241211213024__py3-none-any.whl

liger_kernel/chunked_loss/fused_linear_preference.py

@@ -29,7 +29,7 @@ class LigerFusedLinearPreferenceBase(torch.autograd.Function):
         compute_nll_loss=True,
         compiled=True,
         use_ref_model=False,
-        # TODO: ref input
+        ref_input=None,
         ref_weight=None,
         ref_bias=None,
         **loss_kwargs,
@@ -59,6 +59,7 @@ class LigerFusedLinearPreferenceBase(torch.autograd.Function):
             compute_nll_loss (bool): Whether to compute NLL loss.
             compiled (bool): Whether to use torch compile for chunk accumulation.
             use_ref_model (bool): Whether to use a reference model for the alignment loss.
+            ref_input (torch.Tensor): Reference input tensor. Shape: (batch_size, seq_len, hidden_size).
             ref_weight (torch.Tensor): Reference weight tensor. Shape: (vocab_size, hidden_size).
             ref_bias (torch.Tensor, optional): Reference bias tensor. Shape: (vocab_size,).
             loss_kwargs (dict): Other possible arguments that a loss function might need
@@ -92,6 +93,7 @@ class LigerFusedLinearPreferenceBase(torch.autograd.Function):
             compute_nll_loss=compute_nll_loss,
             full_target=target,
             use_ref_model=use_ref_model,
+            ref_input=ref_input,
             ref_weight=ref_weight,
             ref_bias=ref_bias,
             **loss_kwargs,
@@ -301,6 +303,7 @@ class LigerFusedLinearPreferenceBase(torch.autograd.Function):
         beta=0.1,
         compute_nll_loss=True,
         use_ref_model=False,
+        ref_input=None,
         ref_weight=None,
         ref_bias=None,
         **loss_kwargs,
@@ -319,6 +322,7 @@ class LigerFusedLinearPreferenceBase(torch.autograd.Function):
             beta (float): Weight for the preference loss.
             compute_nll_loss (bool): Whether to compute NLL loss.
             use_ref_model (bool): Whether to use a reference model for the alignment loss.
+            ref_input (torch.Tensor): Reference input tensor. Shape: (2 * chunk_size, sequence_length, hidden_size).
             ref_weight (torch.Tensor): Reference weight tensor. Shape: (vocab_size, hidden_size).
             ref_bias (torch.Tensor, optional): Reference bias tensor. Shape: (vocab_size,).
             loss_kwargs (dict): Additional arguments for the loss function.
@@ -357,7 +361,7 @@ class LigerFusedLinearPreferenceBase(torch.autograd.Function):
                     ref_rejected_logits,
                     ref_chosen_nll_loss,
                 ) = LigerFusedLinearPreferenceBase.chunk_forward(
-                    input_chunk,
+                    ref_input,
                     ref_weight,
                     target_chunk,
                     ref_bias,

{liger_kernel_nightly-0.5.2.dev20241211055800.dist-info → liger_kernel_nightly-0.5.2.dev20241211213024.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: liger_kernel_nightly
-Version: 0.5.2.dev20241211055800
+Version: 0.5.2.dev20241211213024
 Summary: Efficient Triton kernels for LLM Training
 License: BSD 2-CLAUSE LICENSE
         Copyright 2024 LinkedIn Corporation 

{liger_kernel_nightly-0.5.2.dev20241211055800.dist-info → liger_kernel_nightly-0.5.2.dev20241211213024.dist-info}/RECORD

@@ -6,7 +6,7 @@ liger_kernel/chunked_loss/cpo_loss.py,sha256=Qu1Ul2A12sp6CqIT-atPbHWFb_LLtINEA9m
 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/fused_linear_preference.py,sha256=qeRod4MFVttj62uPFhgKAWNNjVrqiEvu5SjZfRnOGzI,15389
 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
@@ -57,9 +57,9 @@ liger_kernel/transformers/trainer/__init__.py,sha256=c4OQVJmhNOloj0JYSEc0j_cQuBb
 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_nightly-0.5.2.dev20241211055800.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
-liger_kernel_nightly-0.5.2.dev20241211055800.dist-info/METADATA,sha256=HSfJ-qbwGmNDotIXkz8RNaQ9h8k2kA6qxAXsPvlv494,20721
-liger_kernel_nightly-0.5.2.dev20241211055800.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
-liger_kernel_nightly-0.5.2.dev20241211055800.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
-liger_kernel_nightly-0.5.2.dev20241211055800.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
-liger_kernel_nightly-0.5.2.dev20241211055800.dist-info/RECORD,,
+liger_kernel_nightly-0.5.2.dev20241211213024.dist-info/LICENSE,sha256=OhzLDHJ0to4a8sodVLELZiCFylZ1NAAYLs-HrjPy0ag,1312
+liger_kernel_nightly-0.5.2.dev20241211213024.dist-info/METADATA,sha256=biU1_vrGRLdmGynYauvf4YfAVHrgN2RtGWe_CNuAD3c,20721
+liger_kernel_nightly-0.5.2.dev20241211213024.dist-info/NOTICE,sha256=njwnoPZLh9AN8SJQzxvCGLHi-8X__AvWRze6joNXIY8,2066
+liger_kernel_nightly-0.5.2.dev20241211213024.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
+liger_kernel_nightly-0.5.2.dev20241211213024.dist-info/top_level.txt,sha256=2eghu4hA3LnkM7ElW92tQ8zegWKgSbeo-k-aGe1YnvY,13
+liger_kernel_nightly-0.5.2.dev20241211213024.dist-info/RECORD,,