support for a new model_type called blueberry

src/transformers/models/auto/configuration_auto.py

@@ -60,6 +60,7 @@
         ("biogpt", "BioGptConfig"),
         ("bit", "BitConfig"),
         ("bitnet", "BitNetConfig"),
+        ("blueberry", "BlueberryConfig"),
         ("blenderbot", "BlenderbotConfig"),
         ("blenderbot-small", "BlenderbotSmallConfig"),
         ("blip", "BlipConfig"),

src/transformers/models/auto/modeling_auto.py

@@ -58,6 +58,7 @@ class _BaseModelWithGenerate(PreTrainedModel, GenerationMixin):
         ("autoformer", "AutoformerModel"),
         ("aya_vision", "AyaVisionModel"),
         ("bamba", "BambaModel"),
+        ("bamba2", "Bamba2Model"),
         ("bark", "BarkModel"),
         ("bart", "BartModel"),
         ("beit", "BeitModel"),
@@ -68,6 +69,7 @@ class _BaseModelWithGenerate(PreTrainedModel, GenerationMixin):
         ("biogpt", "BioGptModel"),
         ("bit", "BitModel"),
         ("bitnet", "BitNetModel"),
+        ("blueberry", "BlueberryModel"),
         ("blenderbot", "BlenderbotModel"),
         ("blenderbot-small", "BlenderbotSmallModel"),
         ("blip", "BlipModel"),
@@ -642,6 +644,7 @@ class _BaseModelWithGenerate(PreTrainedModel, GenerationMixin):
         ("bigbird_pegasus", "BigBirdPegasusForCausalLM"),
         ("biogpt", "BioGptForCausalLM"),
         ("bitnet", "BitNetForCausalLM"),
+        ("blueberry", "BlueberryForCausalLM"),
         ("blenderbot", "BlenderbotForCausalLM"),
         ("blenderbot-small", "BlenderbotSmallForCausalLM"),
         ("bloom", "BloomForCausalLM"),
@@ -1222,6 +1225,7 @@ class _BaseModelWithGenerate(PreTrainedModel, GenerationMixin):
         ("bigbird_pegasus", "BigBirdPegasusForSequenceClassification"),
         ("biogpt", "BioGptForSequenceClassification"),
         ("bloom", "BloomForSequenceClassification"),
+        ("blueberry", "BlueberryForSequenceClassification"),
         ("camembert", "CamembertForSequenceClassification"),
         ("canine", "CanineForSequenceClassification"),
         ("code_llama", "LlamaForSequenceClassification"),

src/transformers/models/auto/tokenization_auto.py

@@ -100,6 +100,7 @@
         ("bigbird_pegasus", ("PegasusTokenizer", "PegasusTokenizerFast" if is_tokenizers_available() else None)),
         ("biogpt", ("BioGptTokenizer", None)),
         ("bitnet", (None, "PreTrainedTokenizerFast" if is_tokenizers_available() else None)),
+        ("blueberry", ("BlueberryTokenizer", "BlueberryTokenizerFast" if is_tokenizers_available() else None)),
         ("blenderbot", ("BlenderbotTokenizer", "BlenderbotTokenizerFast")),
         ("blenderbot-small", ("BlenderbotSmallTokenizer", None)),
         ("blip", ("BertTokenizer", "BertTokenizerFast" if is_tokenizers_available() else None)),

src/transformers/models/blueberry/__init__.py

@@ -0,0 +1,69 @@
+# Copyright 2024 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.
+""" Blueberry model configuration"""
+
+from typing import TYPE_CHECKING
+
+from ...utils import OptionalDependencyNotAvailable, _LazyModule
+
+
+_import_structure = {
+    "configuration_blueberry": [
+        "BLUEBERRY_PRETRAINED_CONFIG_ARCHIVE_MAP",
+        "BlueberryConfig",
+    ],
+}
+
+try:
+    from ...utils import is_torch_available
+    if not is_torch_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_blueberry"] = [
+        "BlueberryForCausalLM",
+        "BlueberryModel",
+        "BlueberryPreTrainedModel",
+        "BlueberryForSequenceClassification",
+    ]
+
+
+if TYPE_CHECKING:
+    from .configuration_blueberry import BLUEBERRY_PRETRAINED_CONFIG_ARCHIVE_MAP, BlueberryConfig
+
+    try:
+        from ...utils import is_torch_available
+        if not is_torch_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_blueberry import (
+            BlueberryForCausalLM,
+            BlueberryForSequenceClassification,
+            BlueberryModel,
+            BlueberryPreTrainedModel,
+        )
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(
+        __name__,
+        globals()["__file__"],
+        _import_structure,
+        module_spec=__spec__,
+        extra_objects={"BLUEBERRY_PRETRAINED_CONFIG_ARCHIVE_MAP": BLUEBERRY_PRETRAINED_CONFIG_ARCHIVE_MAP},
+    )

src/transformers/models/blueberry/configuration_blueberry.py

@@ -0,0 +1,176 @@
+# coding=utf-8
+# Copyright 2024 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.
+""" Blueberry model configuration"""
+
+from ...configuration_utils import PretrainedConfig
+from ...utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+BLUEBERRY_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    # "blueberry": "https://huggingface.co/dloring1988/blueberry/resolve/main/config.json",
+}
+
+
+class BlueberryConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`BlueberryModel`]. It is used to instantiate a
+    blueberry model according to the specified arguments, defining the model architecture. Instantiating a configuration
+    with the defaults will yield a similar configuration to that of the Blueberry
+    [dloring1988/blueberry](https://huggingface.co/dloring1988/blueberry) architecture.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 32000):
+            Vocabulary size of the Blueberry model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`BlueberryModel`]
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 11008):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 32):
+            Number of hidden layers in the Transformer decoder.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        num_key_value_heads (`int`, *optional*, defaults to 32):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details checkout [this
+            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
+            `num_attention_heads`.
+        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
+            The non-linear activation function (function or string) in the decoder.
+        max_position_embeddings (`int`, *optional*, defaults to 2048):
+            The maximum sequence length that this model might ever be used with.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
+            The epsilon used by the rms normalization layers.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        pad_token_id (`int`, *optional*):
+            The id of the padding token.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            The id of the "beginning-of-sequence" token.
+        eos_token_id (`int`, *optional*, defaults to 2):
+            The id of the "end-of-sequence" token.
+        pretraining_tp (`int`, *optional*, defaults to 1):
+            Experimental feature. Tensor parallelism rank used during pretraining with Megatron. Please refer to [this
+            document](https://huggingface.co/docs/transformers/parallelism) to understand more about it. This value is
+            necessary to ensure exact reproducibility of the pretraining results. Please refer to [this
+            issue](https://github.com/pytorch/pytorch/issues/76232).
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether to tie weight embeddings
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        rope_scaling (`Dict`, *optional*):
+            Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
+            strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
+            `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
+            `max_position_embeddings` to the expected new maximum. See the following thread for more information on how
+            these scaling strategies behave:
+            https://www.reddit.com/r/LocalLLaMA/comments/14mrgpr/dynamically_scaled_rope_further_increases/. This is an
+            experimental feature, subject to breaking API changes in future versions.
+        attention_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use a bias in the query, key, value and output projection layers during self-attention.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+        mlp_bias (`bool`, *optional*, defaults to `False`):
+            Whether to use a bias in up_proj, down_proj and gate_proj layers in the MLP module.
+        use_qkv_bias (`bool`, *optional*, defaults to `True`):
+            Whether to use bias in the q_proj, k_proj, and v_proj layers.
+
+    Example:
+
+    ```python
+    >>> from transformers import BlueberryConfig, BlueberryModel
+
+    >>> # Initializing a Blueberry dloring1988/blueberry style configuration
+    >>> configuration = BlueberryConfig()
+
+    >>> # Initializing a model (with random weights) from the dloring1988/blueberry style configuration
+    >>> model = BlueberryModel(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+    model_type = "blueberry"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=32000,
+        hidden_size=4096,
+        intermediate_size=11008,
+        num_hidden_layers=32,
+        num_attention_heads=32,
+        num_key_value_heads=None,
+        hidden_act="silu",
+        max_position_embeddings=2048,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        pad_token_id=None,
+        bos_token_id=1,
+        eos_token_id=2,
+        pretraining_tp=1,
+        tie_word_embeddings=False,
+        rope_theta=10000.0,
+        rope_scaling=None,
+        attention_bias=False,
+        attention_dropout=0.0,
+        mlp_bias=False,
+        use_qkv_bias=True,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+
+        # for backward compatibility
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.pretraining_tp = pretraining_tp
+        self.tie_word_embeddings = tie_word_embeddings
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+        self.mlp_bias = mlp_bias
+        self.use_qkv_bias = use_qkv_bias
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )