OLMo1123 classification.

docs/source/en/_toctree.yml

@@ -514,6 +514,8 @@
         title: Nyströmformer
       - local: model_doc/olmo
         title: OLMo
+      - local: model_doc/olmo_1124
+        title: OLMo November 2024
       - local: model_doc/olmoe
         title: OLMoE
       - local: model_doc/open-llama

docs/source/en/index.md

@@ -240,6 +240,7 @@ Flax), PyTorch, and/or TensorFlow.
 |                        [Nougat](model_doc/nougat)                        |       ✅        |         ✅         |      ✅      |
 |                 [Nyströmformer](model_doc/nystromformer)                 |       ✅        |         ❌         |      ❌      |
 |                          [OLMo](model_doc/olmo)                          |       ✅        |         ❌         |      ❌      |
+|                [OLMo November 2024](model_doc/olmo_1124)                 |       ✅        |         ❌         |      ❌      |
 |                         [OLMoE](model_doc/olmoe)                         |       ✅        |         ❌         |      ❌      |
 |                   [OmDet-Turbo](model_doc/omdet-turbo)                   |       ✅        |         ❌         |      ❌      |
 |                     [OneFormer](model_doc/oneformer)                     |       ✅        |         ❌         |      ❌      |

docs/source/en/model_doc/olmo_1124.md

@@ -0,0 +1,48 @@
+<!--Copyright 2024 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+
+-->
+
+# OLMo November 2024
+
+## Overview
+
+The OLMo November 2024 model was proposed in [<INSERT PAPER NAME HERE>](<INSERT PAPER LINK HERE>) by <INSERT AUTHORS HERE>.
+<INSERT SHORT SUMMARY HERE>
+
+The abstract from the paper is the following:
+
+*<INSERT PAPER ABSTRACT HERE>*
+
+Tips:
+
+<INSERT TIPS ABOUT MODEL HERE>
+
+This model was contributed by [INSERT YOUR HF USERNAME HERE](https://huggingface.co/<INSERT YOUR HF USERNAME HERE>).
+The original code can be found [here](<INSERT LINK TO GITHUB REPO HERE>).
+
+
+## Olmo1124Config
+
+[[autodoc]] Olmo1124Config
+
+## Olmo1124Model
+
+[[autodoc]] Olmo1124Model
+    - forward
+
+## Olmo1124ForCausalLM
+
+[[autodoc]] Olmo1124ForCausalLM
+    - forward

docs/source/en/perf_infer_gpu_one.md

@@ -77,6 +77,7 @@ FlashAttention-2 is currently supported for the following architectures:
 * [Nemotron](https://huggingface.co/docs/transformers/model_doc/nemotron)
 * [NLLB](https://huggingface.co/docs/transformers/model_doc/nllb)
 * [OLMo](https://huggingface.co/docs/transformers/model_doc/olmo#transformers.OlmoModel)
+* [OLMo November 2024](https://huggingface.co/docs/transformers/model_doc/olmo_1124#transformers.Olmo1124Model)
 * [OLMoE](https://huggingface.co/docs/transformers/model_doc/olmoe#transformers.OlmoeModel)
 * [OPT](https://huggingface.co/docs/transformers/model_doc/opt#transformers.OPTModel)
 * [PaliGemma](https://huggingface.co/docs/transformers/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration)
@@ -260,6 +261,7 @@ For now, Transformers supports SDPA inference and training for the following arc
 * [MusicGen Melody](https://huggingface.co/docs/transformers/model_doc/musicgen_melody#transformers.MusicgenMelodyModel)
 * [NLLB](https://huggingface.co/docs/transformers/model_doc/nllb)
 * [OLMo](https://huggingface.co/docs/transformers/model_doc/olmo#transformers.OlmoModel)
+* [OLMo November 2024](https://huggingface.co/docs/transformers/model_doc/olmo_1124#transformers.Olmo1124Model)
 * [OLMoE](https://huggingface.co/docs/transformers/model_doc/olmoe#transformers.OlmoeModel)
 * [OPT](https://huggingface.co/docs/transformers/en/model_doc/opt)
 * [PaliGemma](https://huggingface.co/docs/transformers/model_doc/paligemma#transformers.PaliGemmaForConditionalGeneration)

src/transformers/__init__.py

@@ -620,6 +620,7 @@
     "models.nougat": ["NougatProcessor"],
     "models.nystromformer": ["NystromformerConfig"],
     "models.olmo": ["OlmoConfig"],
+    "models.olmo_1124": ["Olmo1124Config"],
     "models.olmoe": ["OlmoeConfig"],
     "models.omdet_turbo": [
         "OmDetTurboConfig",
@@ -2919,6 +2920,13 @@
             "OlmoPreTrainedModel",
         ]
     )
+    _import_structure["models.olmo_1124"].extend(
+        [
+            "Olmo1124ForCausalLM",
+            "Olmo1124Model",
+            "Olmo1124PreTrainedModel",
+        ]
+    )
     _import_structure["models.olmoe"].extend(
         [
             "OlmoeForCausalLM",
@@ -5506,6 +5514,7 @@
         NystromformerConfig,
     )
     from .models.olmo import OlmoConfig
+    from .models.olmo_1124 import Olmo1124Config
     from .models.olmoe import OlmoeConfig
     from .models.omdet_turbo import (
         OmDetTurboConfig,
@@ -7523,6 +7532,11 @@
             OlmoModel,
             OlmoPreTrainedModel,
         )
+        from .models.olmo_1124 import (
+            Olmo1124ForCausalLM,
+            Olmo1124Model,
+            Olmo1124PreTrainedModel,
+        )
         from .models.olmoe import (
             OlmoeForCausalLM,
             OlmoeModel,

src/transformers/models/__init__.py

@@ -177,6 +177,7 @@
     nougat,
     nystromformer,
     olmo,
+    olmo_1124,
     olmoe,
     omdet_turbo,
     oneformer,

src/transformers/models/auto/configuration_auto.py

@@ -195,6 +195,7 @@
         ("nougat", "VisionEncoderDecoderConfig"),
         ("nystromformer", "NystromformerConfig"),
         ("olmo", "OlmoConfig"),
+        ("olmo_1124", "Olmo1124Config"),
         ("olmoe", "OlmoeConfig"),
         ("omdet-turbo", "OmDetTurboConfig"),
         ("oneformer", "OneFormerConfig"),
@@ -510,6 +511,7 @@
         ("nougat", "Nougat"),
         ("nystromformer", "Nyströmformer"),
         ("olmo", "OLMo"),
+        ("olmo_1124", "OLMo November 2024"),
         ("olmoe", "OLMoE"),
         ("omdet-turbo", "OmDet-Turbo"),
         ("oneformer", "OneFormer"),

src/transformers/models/auto/modeling_auto.py

@@ -184,6 +184,7 @@
         ("nllb-moe", "NllbMoeModel"),
         ("nystromformer", "NystromformerModel"),
         ("olmo", "OlmoModel"),
+        ("olmo_1124", "Olmo1124Model"),
         ("olmoe", "OlmoeModel"),
         ("omdet-turbo", "OmDetTurboForObjectDetection"),
         ("oneformer", "OneFormerModel"),
@@ -516,6 +517,7 @@
         ("mvp", "MvpForCausalLM"),
         ("nemotron", "NemotronForCausalLM"),
         ("olmo", "OlmoForCausalLM"),
+        ("olmo_1124", "Olmo1124ForCausalLM"),
         ("olmoe", "OlmoeForCausalLM"),
         ("open-llama", "OpenLlamaForCausalLM"),
         ("openai-gpt", "OpenAIGPTLMHeadModel"),

src/transformers/models/auto/tokenization_auto.py

@@ -348,6 +348,7 @@
                 ),
             ),
             ("olmo", (None, "GPTNeoXTokenizerFast" if is_tokenizers_available() else None)),
+            ("olmo_1124", (None, "GPTNeoXTokenizerFast" if is_tokenizers_available() else None)),
             ("olmoe", (None, "GPTNeoXTokenizerFast" if is_tokenizers_available() else None)),
             (
                 "omdet-turbo",

src/transformers/models/olmo/modeling_olmo.py

@@ -26,6 +26,8 @@
 import torch.nn.functional as F
 import torch.utils.checkpoint
 from torch import nn
+from torch.nn import MSELoss, CrossEntropyLoss, BCEWithLogitsLoss
+
 
 from ...activations import ACT2FN
 from ...cache_utils import Cache, DynamicCache, StaticCache
@@ -34,6 +36,7 @@
 from ...modeling_outputs import (
     BaseModelOutputWithPast,
     CausalLMOutputWithPast,
+    SequenceClassifierOutputWithPast,
 )
 from ...modeling_utils import PreTrainedModel
 from ...pytorch_utils import ALL_LAYERNORM_LAYERS
@@ -1140,3 +1143,110 @@ def forward(
             hidden_states=outputs.hidden_states,
             attentions=outputs.attentions,
         )
+
+
+class OlmoForSequenceClassification(OlmoPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.model = OlmoModel(config)
+        self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Union[List[torch.FloatTensor]]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, SequenceClassifierOutputWithPast]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.model(
+            input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        hidden_states = transformer_outputs[0]
+        logits = self.score(hidden_states)
+
+        if input_ids is not None:
+            batch_size = input_ids.shape[0]
+        else:
+            batch_size = inputs_embeds.shape[0]
+
+        if self.config.pad_token_id is None and batch_size != 1:
+            raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
+        if self.config.pad_token_id is None:
+            sequence_lengths = -1
+        else:
+            if input_ids is not None:
+                # if no pad token found, use modulo instead of reverse indexing for ONNX compatibility
+                sequence_lengths = torch.eq(input_ids, self.config.pad_token_id).int().argmax(-1) - 1
+                sequence_lengths = sequence_lengths % input_ids.shape[-1]
+                sequence_lengths = sequence_lengths.to(logits.device)
+            else:
+                sequence_lengths = -1
+
+        pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths]
+
+        loss = None
+        if labels is not None:
+            labels = labels.to(logits.device)
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(pooled_logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(pooled_logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(pooled_logits, labels)
+        if not return_dict:
+            output = (pooled_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=pooled_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )

src/transformers/models/olmo_1124/__init__.py

@@ -0,0 +1,57 @@
+# Copyright 2024 EleutherAI and The HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import TYPE_CHECKING
+
+from ...utils import (
+    OptionalDependencyNotAvailable,
+    _LazyModule,
+    is_torch_available,
+)
+
+
+_import_structure = {
+    "configuration_olmo_1124": ["Olmo1124Config"],
+}
+
+try:
+    if not is_torch_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_olmo_1124"] = [
+        "Olmo1124ForCausalLM",
+        "Olmo1124Model",
+        "Olmo1124PreTrainedModel",
+    ]
+
+if TYPE_CHECKING:
+    from .configuration_olmo_1124 import Olmo1124Config
+
+    try:
+        if not is_torch_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_olmo_1124 import (
+            Olmo1124ForCausalLM,
+            Olmo1124Model,
+            Olmo1124PreTrainedModel,
+        )
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)