Skip to content

Check model inputs - hidden states #40994

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Oct 6, 2025
Merged

Check model inputs - hidden states #40994

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
9ee3f43
update all models
zucchini-nlp Sep 18, 2025
63bb723
fix copies
zucchini-nlp Sep 18, 2025
6419997
merge main
zucchini-nlp Sep 24, 2025
883008b
skip aria tests
zucchini-nlp Sep 24, 2025
0db1f50
update other models
zucchini-nlp Sep 24, 2025
e5d6d59
skip should be in test, not tester
zucchini-nlp Sep 24, 2025
c0c58e1
i think this is more descriptive as a name
zucchini-nlp Oct 6, 2025
310a94c
Merge branch 'main' into check-model-inputs-hidden
zucchini-nlp Oct 6, 2025
68a7b7e
find and replace for new models
zucchini-nlp Oct 6, 2025
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion examples/modular-transformers/modeling_dummy_bert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -657,7 +657,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion examples/modular-transformers/modeling_roberta.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -660,7 +660,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion examples/modular-transformers/modeling_super.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -323,7 +323,7 @@ def __init__(self, config: SuperConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/aimv2/modeling_aimv2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -444,7 +444,7 @@ def get_input_embeddings(self) -> nn.Module:
return self.embeddings.patch_embed

@deprecate_kwarg("attention_mask", version="v4.58.0")
@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -520,7 +520,7 @@ def get_input_embeddings(self) -> nn.Module:
def set_input_embeddings(self, value):
self.embeddings.token_embedding = value

@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/aimv2/modular_aimv2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -488,7 +488,7 @@ def get_input_embeddings(self) -> nn.Module:
return self.embeddings.patch_embed

@deprecate_kwarg("attention_mask", version="v4.58.0")
@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -564,7 +564,7 @@ def get_input_embeddings(self) -> nn.Module:
def set_input_embeddings(self, value):
self.embeddings.token_embedding = value

@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/albert/modeling_albert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -457,7 +457,7 @@ def _prune_heads(self, heads_to_prune: dict[int, list[int]]) -> None:
inner_group_idx = int(layer - group_idx * self.config.inner_group_num)
self.encoder.albert_layer_groups[group_idx].albert_layers[inner_group_idx].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/apertus/modeling_apertus.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -339,7 +339,7 @@ def __init__(self, config: ApertusConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/arcee/modeling_arcee.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -344,7 +344,7 @@ def __init__(self, config: ArceeConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/aria/modeling_aria.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -669,7 +669,7 @@ def __init__(self, config: AriaTextConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion ...ansformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -363,7 +363,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/aya_vision/modeling_aya_vision.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -263,7 +263,7 @@ def get_placeholder_mask(
)
return special_image_mask

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/aya_vision/modular_aya_vision.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -162,7 +162,7 @@ def get_image_features(
image_features = self.multi_modal_projector(selected_image_feature)
return image_features

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/bert/modeling_bert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -739,7 +739,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/bert_generation/modeling_bert_generation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -600,7 +600,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/bitnet/modeling_bitnet.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -343,7 +343,7 @@ def __init__(self, config: BitNetConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/blip/modeling_blip.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -499,7 +499,7 @@ def __init__(self, config: BlipVisionConfig):

self.post_init()

@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/blip_2/modeling_blip_2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -490,7 +490,7 @@ def __init__(self, config: Blip2VisionConfig):

self.post_init()

@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -989,7 +989,7 @@ def get_extended_attention_mask(
extended_attention_mask = (1.0 - extended_attention_mask) * -10000.0
return extended_attention_mask

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/blt/modeling_blt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -577,7 +577,7 @@ def __init__(self, config: BltLocalDecoderConfig):

self.post_init()

@check_model_inputs
@check_model_inputs()
def forward(
self,
input_ids: Optional[torch.LongTensor] = None,
Expand Down Expand Up @@ -1047,7 +1047,7 @@ def __init__(self, config: BltConfig):
self.patcher = None
self.post_init()

@check_model_inputs
@check_model_inputs()
def forward(
self,
input_ids: Optional[torch.LongTensor] = None,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/blt/modular_blt.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -536,7 +536,7 @@ def __init__(self, config: BltLocalDecoderConfig):

self.post_init()

@check_model_inputs
@check_model_inputs()
def forward(
self,
input_ids: Optional[torch.LongTensor] = None,
Expand Down Expand Up @@ -799,7 +799,7 @@ def __init__(self, config: BltConfig):
self.patcher = None
self.post_init()

@check_model_inputs
@check_model_inputs()
def forward(
self,
input_ids: Optional[torch.LongTensor] = None,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/camembert/modeling_camembert.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -719,7 +719,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/cohere/modeling_cohere.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -376,7 +376,7 @@ def __init__(self, config: CohereConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/cohere2/modeling_cohere2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -351,7 +351,7 @@ def __init__(self, config: Cohere2Config):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/cohere2_vision/modeling_cohere2_vision.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -213,7 +213,7 @@ def get_placeholder_mask(
)
return special_image_mask

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -306,7 +306,7 @@ def vision_tower(self):
def multi_modal_projector(self):
return self.model.multi_modal_projector

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/cohere2_vision/modular_cohere2_vision.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,7 +109,7 @@ def get_image_features(self, pixel_values: torch.FloatTensor):
image_features = self.multi_modal_projector(selected_image_feature)
return image_features

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -162,7 +162,7 @@ class Cohere2VisionForConditionalGeneration(AyaVisionForConditionalGeneration):
def get_image_features(self, pixel_values: torch.FloatTensor):
return self.model.get_image_features(pixel_values=pixel_values)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/csm/modeling_csm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -409,7 +409,7 @@ def __init__(self, config):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -662,7 +662,7 @@ def __init__(self, config):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/csm/modular_csm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -156,7 +156,7 @@ def __init__(self, config):
self.embed_tokens = nn.Embedding((config.num_codebooks * config.vocab_size), config.backbone_hidden_size)
self.inputs_embeds_projector = nn.Linear(config.backbone_hidden_size, config.hidden_size, bias=False)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -395,7 +395,7 @@ def __init__(self, config):
super().__init__(config)
self.embed_tokens = CsmBackboneModelEmbeddings(config)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(self, **super_kwargs):
r"""
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/data2vec/modeling_data2vec_text.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -679,7 +679,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/dbrx/modeling_dbrx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -489,7 +489,7 @@ def get_input_embeddings(self) -> nn.Embedding:
def set_input_embeddings(self, value: nn.Embedding):
self.wte = value

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/dbrx/modular_dbrx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -388,7 +388,7 @@ def get_input_embeddings(self) -> nn.Embedding:
def set_input_embeddings(self, value: nn.Embedding):
self.wte = value

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/deepseek_v2/modeling_deepseek_v2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -459,7 +459,7 @@ def __init__(self, config: DeepseekV2Config):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/deepseek_v3/modeling_deepseek_v3.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -539,7 +539,7 @@ def __init__(self, config: DeepseekV3Config):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/deit/modeling_deit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -438,7 +438,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/diffllama/modeling_diffllama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -608,7 +608,7 @@ def __init__(self, config: DiffLlamaConfig):
# Initialize weights and apply final processing
self.post_init()

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self,
Expand Down
4 changes: 2 additions & 2 deletions src/transformers/models/dinov2/modeling_dinov2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -490,7 +490,7 @@ class PreTrainedModel
for layer, heads in heads_to_prune.items():
self.encoder.layer[layer].attention.prune_heads(heads)

@check_model_inputs
@check_model_inputs(tie_last_hidden_states=False)
@auto_docstring
def forward(
self,
Expand Down Expand Up @@ -602,7 +602,7 @@ def __init__(self, config):
def get_input_embeddings(self) -> Dinov2PatchEmbeddings:
return self.embeddings.patch_embeddings

@check_model_inputs
@check_model_inputs()
@auto_docstring
def forward(
self, pixel_values: torch.Tensor, output_hidden_states: Optional[bool] = None, **kwargs
Expand Down
Loading