Phi-3 (#30423)

* chore(root): Initial commit of Phi-3 files.

* fix(root): Fixes Phi-3 missing on readme.

* fix(root): Ensures files are consistent.

* fix(phi3): Fixes unit tests.

* fix(tests): Fixes style of phi-3 test file.

* chore(tests): Adds integration tests for Phi-3.

* fix(phi3): Removes additional flash-attention usage, .e.g, swiglu and rmsnorm.

* fix(phi3): Fixes incorrect docstrings.

* fix(phi3): Fixes docstring typos.

* fix(phi3): Adds support for Su and Yarn embeddings.

* fix(phi3): Improves according first batch of reviews.

* fix(phi3): Uses up_states instead of y in Phi3MLP.

* fix(phi3): Uses gemma rotary embedding to support torch.compile.

* fix(phi3): Improves how rotary embedding classes are defined.

* fix(phi3): Fixes inv_freq not being re-computed for extended RoPE.

* fix(phi3): Adds last suggestions to modeling file.

* fix(phi3): Splits inv_freq calculation in two lines.

docs/source/en/_toctree.yml

@@ -460,6 +460,8 @@
         title: Persimmon
       - local: model_doc/phi
         title: Phi
+      - local: model_doc/phi3
+        title: Phi-3
       - local: model_doc/phobert
         title: PhoBERT
       - local: model_doc/plbart

docs/source/en/index.md

@@ -236,6 +236,7 @@ Flax), PyTorch, and/or TensorFlow.
 |                     [Perceiver](model_doc/perceiver)                     |       ✅        |         ❌         |      ❌      |
 |                     [Persimmon](model_doc/persimmon)                     |       ✅        |         ❌         |      ❌      |
 |                           [Phi](model_doc/phi)                           |       ✅        |         ❌         |      ❌      |
+|                          [Phi3](model_doc/phi3)                          |       ✅        |         ❌         |      ❌      |
 |                       [PhoBERT](model_doc/phobert)                       |       ✅        |         ✅         |      ✅      |
 |                    [Pix2Struct](model_doc/pix2struct)                    |       ✅        |         ❌         |      ❌      |
 |                        [PLBart](model_doc/plbart)                        |       ✅        |         ❌         |      ❌      |

docs/source/en/model_doc/phi3.md

+<!--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.
+
+-->
+
+# Phi-3
+
+## Overview
+
+The Phi-3 model was proposed in [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://arxiv.org/abs/2404.14219) by Microsoft.
+
+### Summary
+
+The abstract from the Phi-3 paper is the following:
+
+We introduce phi-3-mini, a 3.8 billion parameter language model trained on 3.3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and GPT-3.5 (e.g., phi-3-mini achieves 69% on MMLU and 8.38 on MT-bench), despite being small enough to be deployed on a phone. The innovation lies entirely in our dataset for training, a scaled-up version of the one used for phi-2, composed of heavily filtered web data and synthetic data. The model is also further aligned for robustness, safety, and chat format. We also provide some initial parameter-scaling results with a 7B and 14B models trained for 4.8T tokens, called phi-3-small and phi-3-medium, both significantly more capable than phi-3-mini (e.g., respectively 75% and 78% on MMLU, and 8.7 and 8.9 on MT-bench).
+
+The original code for Phi-3 can be found [here](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct).
+
+## Usage tips
+
+- This model is very similar to `Llama` with the main difference of [`Phi3SuScaledRotaryEmbedding`] and [`Phi3YarnScaledRotaryEmbedding`], where they are used to extend the context of the rotary embeddings. The query, key and values are fused, and the MLP's up and gate projection layers are also fused.
+- The tokenizer used for this model is identical to the [`LlamaTokenizer`], with the exception of additional tokens.
+
+## How to use Phi-3
+
+<Tip warning={true}>
+
+Phi-3 has been integrated in the development version (4.40.0.dev) of `transformers`. Until the official version is released through `pip`, ensure that you are doing one of the following:
+
+* When loading the model, ensure that `trust_remote_code=True` is passed as an argument of the `from_pretrained()` function.
+
+* Update your local `transformers` to the development version: `pip uninstall -y transformers && pip install git+https://github.com/huggingface/transformers`. The previous command is an alternative to cloning and installing from the source.
+
+</Tip>
+
+```python
+>>> from transformers import AutoModelForCausalLM, AutoTokenizer
+
+>>> model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+
+>>> messages = [{"role": "system", "content": "You are a helpful digital assistant. Please provide safe, ethical and accurate information to the user."},{"role": "user", "content": "Can you provide ways to eat combinations of bananas and dragonfruits?"}]
+>>> inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
+
+>>> outputs = model.generate(inputs, max_new_tokens=32)
+>>> text = tokenizer.batch_decode(outputs)[0]
+>>> print(text)
+<s><|system|> You are a helpful digital assistant. Please provide safe, ethical and accurate information to the user.<|end|><|user|> Can you provide ways to eat combinations of bananas and dragonfruits?<|end|><|assistant|> Absolutely! Bananas and dragonfruits are both delicious fruits that can be combined in various ways to create tasty and nutrit
+```
+
+## Phi3Config
+
+[[autodoc]] Phi3Config
+
+<frameworkcontent>
+<pt>
+
+## Phi3Model
+
+[[autodoc]] Phi3Model
+    - forward
+
+## Phi3ForCausalLM
+
+[[autodoc]] Phi3ForCausalLM
+    - forward
+    - generate
+
+## Phi3ForSequenceClassification
+
+[[autodoc]] Phi3ForSequenceClassification
+    - forward
+
+## Phi3ForTokenClassification
+
+[[autodoc]] Phi3ForTokenClassification
+    - forward
+
+</pt>
+</frameworkcontent>

docs/source/en/perf_infer_gpu_one.md

@@ -65,6 +65,7 @@ FlashAttention-2 is currently supported for the following architectures:
 * [OLMo](https://huggingface.co/docs/transformers/model_doc/olmo#transformers.OlmoModel)
 * [OPT](https://huggingface.co/docs/transformers/model_doc/opt#transformers.OPTModel)
 * [Phi](https://huggingface.co/docs/transformers/model_doc/phi#transformers.PhiModel)
+* [Phi3](https://huggingface.co/docs/transformers/model_doc/phi3#transformers.Phi3Model)
 * [StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm#transformers.StableLmModel)
 * [Starcoder2](https://huggingface.co/docs/transformers/model_doc/starcoder2#transformers.Starcoder2Model)
 * [Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2#transformers.Qwen2Model)

src/transformers/__init__.py

@@ -709,6 +709,7 @@ _import_structure = {
     ],
     "models.persimmon": ["PERSIMMON_PRETRAINED_CONFIG_ARCHIVE_MAP", "PersimmonConfig"],
     "models.phi": ["PHI_PRETRAINED_CONFIG_ARCHIVE_MAP", "PhiConfig"],
+    "models.phi3": ["PHI3_PRETRAINED_CONFIG_ARCHIVE_MAP", "Phi3Config"],
     "models.phobert": ["PhobertTokenizer"],
     "models.pix2struct": [
         "PIX2STRUCT_PRETRAINED_CONFIG_ARCHIVE_MAP",
@@ -3057,6 +3058,16 @@ else:
             "PhiPreTrainedModel",
         ]
     )
+    _import_structure["models.phi3"].extend(
+        [
+            "PHI3_PRETRAINED_MODEL_ARCHIVE_LIST",
+            "Phi3ForCausalLM",
+            "Phi3ForSequenceClassification",
+            "Phi3ForTokenClassification",
+            "Phi3Model",
+            "Phi3PreTrainedModel",
+        ]
+    )
     _import_structure["models.pix2struct"].extend(
         [
             "PIX2STRUCT_PRETRAINED_MODEL_ARCHIVE_LIST",
@@ -5669,6 +5680,7 @@ if TYPE_CHECKING:
         PersimmonConfig,
     )
     from .models.phi import PHI_PRETRAINED_CONFIG_ARCHIVE_MAP, PhiConfig
+    from .models.phi3 import PHI3_PRETRAINED_CONFIG_ARCHIVE_MAP, Phi3Config
     from .models.phobert import PhobertTokenizer
     from .models.pix2struct import (
         PIX2STRUCT_PRETRAINED_CONFIG_ARCHIVE_MAP,
@@ -7715,6 +7727,14 @@ if TYPE_CHECKING:
             PhiModel,
             PhiPreTrainedModel,
         )
+        from .models.phi3 import (
+            PHI3_PRETRAINED_MODEL_ARCHIVE_LIST,
+            Phi3ForCausalLM,
+            Phi3ForSequenceClassification,
+            Phi3ForTokenClassification,
+            Phi3Model,
+            Phi3PreTrainedModel,
+        )
         from .models.pix2struct import (
             PIX2STRUCT_PRETRAINED_MODEL_ARCHIVE_LIST,
             Pix2StructForConditionalGeneration,

src/transformers/models/__init__.py

@@ -179,6 +179,7 @@ from . import (
     perceiver,
     persimmon,
     phi,
+    phi3,
     phobert,
     pix2struct,
     plbart,

src/transformers/models/auto/configuration_auto.py

@@ -191,6 +191,7 @@ CONFIG_MAPPING_NAMES = OrderedDict(
         ("perceiver", "PerceiverConfig"),
         ("persimmon", "PersimmonConfig"),
         ("phi", "PhiConfig"),
+        ("phi3", "Phi3Config"),
         ("pix2struct", "Pix2StructConfig"),
         ("plbart", "PLBartConfig"),
         ("poolformer", "PoolFormerConfig"),
@@ -471,6 +472,7 @@ MODEL_NAMES_MAPPING = OrderedDict(
         ("perceiver", "Perceiver"),
         ("persimmon", "Persimmon"),
         ("phi", "Phi"),
+        ("phi3", "Phi3"),
         ("phobert", "PhoBERT"),
         ("pix2struct", "Pix2Struct"),
         ("plbart", "PLBart"),

src/transformers/models/auto/modeling_auto.py

@@ -180,6 +180,7 @@ MODEL_MAPPING_NAMES = OrderedDict(
         ("perceiver", "PerceiverModel"),
         ("persimmon", "PersimmonModel"),
         ("phi", "PhiModel"),
+        ("phi3", "Phi3Model"),
         ("plbart", "PLBartModel"),
         ("poolformer", "PoolFormerModel"),
         ("prophetnet", "ProphetNetModel"),
@@ -474,6 +475,7 @@ MODEL_FOR_CAUSAL_LM_MAPPING_NAMES = OrderedDict(
         ("pegasus", "PegasusForCausalLM"),
         ("persimmon", "PersimmonForCausalLM"),
         ("phi", "PhiForCausalLM"),
+        ("phi3", "Phi3ForCausalLM"),
         ("plbart", "PLBartForCausalLM"),
         ("prophetnet", "ProphetNetForCausalLM"),
         ("qdqbert", "QDQBertLMHeadModel"),
@@ -884,6 +886,7 @@ MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING_NAMES = OrderedDict(
         ("perceiver", "PerceiverForSequenceClassification"),
         ("persimmon", "PersimmonForSequenceClassification"),
         ("phi", "PhiForSequenceClassification"),
+        ("phi3", "Phi3ForSequenceClassification"),
         ("plbart", "PLBartForSequenceClassification"),
         ("qdqbert", "QDQBertForSequenceClassification"),
         ("qwen2", "Qwen2ForSequenceClassification"),
@@ -1049,6 +1052,7 @@ MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMES = OrderedDict(
         ("nezha", "NezhaForTokenClassification"),
         ("nystromformer", "NystromformerForTokenClassification"),
         ("phi", "PhiForTokenClassification"),
+        ("phi3", "Phi3ForTokenClassification"),
         ("qdqbert", "QDQBertForTokenClassification"),
         ("rembert", "RemBertForTokenClassification"),
         ("roberta", "RobertaForTokenClassification"),

src/transformers/models/auto/tokenization_auto.py

@@ -353,6 +353,7 @@ else:
                 ),
             ),
             ("phi", ("CodeGenTokenizer", "CodeGenTokenizerFast" if is_tokenizers_available() else None)),
+            ("phi3", ("LlamaTokenizer", "LlamaTokenizerFast" if is_tokenizers_available() else None)),
             ("phobert", ("PhobertTokenizer", None)),
             ("pix2struct", ("T5Tokenizer", "T5TokenizerFast" if is_tokenizers_available() else None)),
             ("plbart", ("PLBartTokenizer" if is_sentencepiece_available() else None, None)),

src/transformers/models/phi3/__init__.py

+# Copyright 2024 Microsoft 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_sentencepiece_available,
+    is_tokenizers_available,
+    is_torch_available,
+)
+
+
+_import_structure = {
+    "configuration_phi3": ["PHI3_PRETRAINED_CONFIG_ARCHIVE_MAP", "Phi3Config"],
+}
+
+try:
+    if not is_torch_available():
+        raise OptionalDependencyNotAvailable()
+except OptionalDependencyNotAvailable:
+    pass
+else:
+    _import_structure["modeling_phi3"] = [
+        "PHI3_PRETRAINED_MODEL_ARCHIVE_LIST",
+        "Phi3PreTrainedModel",
+        "Phi3Model",
+        "Phi3ForCausalLM",
+        "Phi3ForSequenceClassification",
+        "Phi3ForTokenClassification",
+    ]
+
+
+if TYPE_CHECKING:
+    from .configuration_phi3 import PHI3_PRETRAINED_CONFIG_ARCHIVE_MAP, Phi3Config
+
+    try:
+        if not is_torch_available():
+            raise OptionalDependencyNotAvailable()
+    except OptionalDependencyNotAvailable:
+        pass
+    else:
+        from .modeling_phi3 import (
+            PHI3_PRETRAINED_MODEL_ARCHIVE_LIST,
+            Phi3ForCausalLM,
+            Phi3ForSequenceClassification,
+            Phi3ForTokenClassification,
+            Phi3Model,
+            Phi3PreTrainedModel,
+        )
+
+
+else:
+    import sys
+
+    sys.modules[__name__] = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)

src/transformers/models/phi3/configuration_phi3.py

+# coding=utf-8
+# Copyright 2024 Microsoft 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.
+
+""" Phi-3 model configuration"""
+
+
+from ...configuration_utils import PretrainedConfig
+from ...utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+PHI3_PRETRAINED_CONFIG_ARCHIVE_MAP = {
+    "microsoft/Phi-3-mini-4k-instruct": "https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/resolve/main/config.json",
+    "microsoft/Phi-3-mini-128k-instruct": "https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/resolve/main/config.json",
+}
+
+
+class Phi3Config(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`Phi3Model`]. It is used to instantiate a Phi-3
+    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
+    [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct).
+
+    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 32064):
+            Vocabulary size of the Phi-3 model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`Phi3Model`].
+        hidden_size (`int`, *optional*, defaults to 3072):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 8192):
+            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*):
+            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`.
+        resid_pdrop (`float`, *optional*, defaults to 0.0):
+            Dropout probability for mlp outputs.
+        embd_pdrop (`int`, *optional*, defaults to 0.0):
+            The dropout ratio for the embeddings.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio after computing the attention scores.
+        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 4096):
+            The maximum sequence length that this model might ever be used with.
+        original_max_position_embeddings (`int`, *optional*, defaults to 4096):
+            The maximum sequence length that this model was trained with. This is used to determine the size of the
+            original RoPE embeddings when using long scaling.
+        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-05):
+            The epsilon value used for the RMSNorm.
+        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`. Whether to tie weight embeddings or not.
+        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*):
+            The scaling strategy for the RoPE embeddings. If `None`, no scaling is applied. If a dictionary, it must
+            contain the following keys: `type`, `short_factor` and `long_factor`. The `type` must be either `su` or `yarn` and
+            the `short_factor` and `long_factor` must be lists of numbers with the same length as the hidden size
+            divided by the number of attention heads divided by 2.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            The id of the "beginning-of-sequence" token.
+        eos_token_id (`int`, *optional*, defaults to 32000):
+            The id of the "end-of-sequence" token.
+        pad_token_id (`int`, *optional*, defaults to 32000):
+            The id of the padding token.
+        sliding_window (`int`, *optional*):
+            Sliding window attention window size. If `None`, no sliding window is applied.
+
+    Example:
+
+    ```python
+    >>> from transformers import Phi3Model, Phi3Config
+
+    >>> # Initializing a Phi-3 style configuration
+    >>> configuration = Phi3Config.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+
+    >>> # Initializing a model from the configuration
+    >>> model = Phi3Model(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "phi3"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=32064,
+        hidden_size=3072,
+        intermediate_size=8192,
+        num_hidden_layers=32,
+        num_attention_heads=32,
+        num_key_value_heads=None,
+        resid_pdrop=0.0,
+        embd_pdrop=0.0,
+        attention_dropout=0.0,
+        hidden_act="silu",
+        max_position_embeddings=4096,
+        original_max_position_embeddings=4096,
+        initializer_range=0.02,
+        rms_norm_eps=1e-5,
+        use_cache=True,
+        tie_word_embeddings=False,
+        rope_theta=10000.0,
+        rope_scaling=None,
+        bos_token_id=1,
+        eos_token_id=32000,
+        pad_token_id=32000,
+        sliding_window=None,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+
+        self.num_key_value_heads = num_key_value_heads
+        self.resid_pdrop = resid_pdrop
+        self.embd_pdrop = embd_pdrop
+        self.attention_dropout = attention_dropout
+        self.hidden_act = hidden_act
+        self.max_position_embeddings = max_position_embeddings
+        self.original_max_position_embeddings = original_max_position_embeddings
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self._rope_scaling_validation()
+        self.sliding_window = sliding_window
+
+        super().__init__(
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            pad_token_id=pad_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+
+    def _rope_scaling_validation(self):
+        """
+        Validate the `rope_scaling` configuration.
+        """
+        if self.rope_scaling is None:
+            return
+
+        if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 3:
+            raise ValueError(
+                "`rope_scaling` must be a dictionary with three fields, `type`, `short_factor` and `long_factor`, "
+                f"got {self.rope_scaling}"
+            )
+        rope_scaling_type = self.rope_scaling.get("type", None)
+        rope_scaling_short_factor = self.rope_scaling.get("short_factor", None)
+        rope_scaling_long_factor = self.rope_scaling.get("long_factor", None)
+        if rope_scaling_type is None or rope_scaling_type not in ["su", "yarn"]:
+            raise ValueError(f"`rope_scaling`'s type field must be one of ['su', 'yarn'], got {rope_scaling_type}")
+        if not (
+            isinstance(rope_scaling_short_factor, list)
+            and all(isinstance(x, (int, float)) for x in rope_scaling_short_factor)
+        ):
+            raise ValueError(
+                f"`rope_scaling`'s short_factor field must be a list of numbers, got {rope_scaling_short_factor}"
+            )
+        if not len(rope_scaling_short_factor) == self.hidden_size // self.num_attention_heads // 2:
+            raise ValueError(
+                f"`rope_scaling`'s short_factor field must have length {self.hidden_size // self.num_attention_heads // 2}, got {len(rope_scaling_short_factor)}"
+            )
+        if not (
+            isinstance(rope_scaling_long_factor, list)
+            and all(isinstance(x, (int, float)) for x in rope_scaling_long_factor)
+        ):
+            raise ValueError(
+                f"`rope_scaling`'s long_factor field must be a list of numbers, got {rope_scaling_long_factor}"
+            )
+        if not len(rope_scaling_long_factor) == self.hidden_size // self.num_attention_heads // 2:
+            raise ValueError(
+                f"`rope_scaling`'s long_factor field must have length {self.hidden_size // self.num_attention_heads // 2}, got {len(rope_scaling_long_factor)}"
+            )

src/transformers/utils/dummy_pt_objects.py

@@ -6752,6 +6752,44 @@ class PhiPreTrainedModel(metaclass=DummyObject):
         requires_backends(self, ["torch"])
 
 
+PHI3_PRETRAINED_MODEL_ARCHIVE_LIST = None
+
+
+class Phi3ForCausalLM(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class Phi3ForSequenceClassification(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class Phi3ForTokenClassification(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class Phi3Model(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
+class Phi3PreTrainedModel(metaclass=DummyObject):
+    _backends = ["torch"]
+
+    def __init__(self, *args, **kwargs):
+        requires_backends(self, ["torch"])
+
+
 PIX2STRUCT_PRETRAINED_MODEL_ARCHIVE_LIST = None