gpt_neo

`mindnlp.transformers.models.gpt_neo.configuration_gpt_neo` ¶

GPT Neo model configuration

`mindnlp.transformers.models.gpt_neo.configuration_gpt_neo.GPTNeoConfig` ¶

Bases: PretrainedConfig

This is the configuration class to store the configuration of a [GPTNeoModel]. It is used to instantiate a GPT Neo 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 GPTNeo EleutherAI/gpt-neo-1.3B architecture.

Configuration objects inherit from [PretrainedConfig] and can be used to control the model outputs. Read the documentation from [PretrainedConfig] for more information.

PARAMETER	DESCRIPTION
`vocab_size`	Vocabulary size of the GPT Neo model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`GPTNeoModel`]. Vocabulary size of the model. Defines the different tokens that can be represented by the inputs_ids passed to the forward method of [`GPTNeoModel`]. TYPE: `int`, optional, defaults to 50257 DEFAULT: `50257`
`max_position_embeddings`	The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048). TYPE: `int`, optional, defaults to 2048 DEFAULT: `2048`
`hidden_size`	Dimensionality of the encoder layers and the pooler layer. TYPE: `int`, optional, defaults to 2048 DEFAULT: `2048`
`num_layers`	Number of hidden layers in the Transformer encoder. TYPE: `int`, optional, defaults to 24 DEFAULT: `24`
`attention_types`	The type of attention for each layer in a `List` of the following format `[[["attention_type"], num_layerss]]` e.g. for a 24 layer model `[[["global"], 24]]` or `[[["global", "local"], 12]]` Choose the value of `attention_type` from `["global", "local"]` TYPE: `List`, optional, defaults to `[[['global', 'local'], 12]]` DEFAULT: `[[['global', 'local'], 12]]`
`num_heads`	Number of attention heads for each attention layer in the Transformer encoder. TYPE: `int`, optional, defaults to 16 DEFAULT: `16`
`intermediate_size`	Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder. TYPE: `int`, optional, defaults to 8192 DEFAULT: `None`
`window_size`	The size of the sliding window for local attention. TYPE: `int`, optional, defaults to 256 DEFAULT: `256`
`activation_function`	The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`, `"relu"`, `"selu"` and `"gelu_new"` are supported. TYPE: `str` or `function`, optional, defaults to `"gelu_new"` DEFAULT: `'gelu_new'`
`resid_dropout`	Residual dropout used in the attention pattern. TYPE: `float`, optional, defaults to 0.0 DEFAULT: `0.0`
`embed_dropout`	The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. TYPE: `float`, optional, defaults to 0.0 DEFAULT: `0.0`
`attention_dropout`	The dropout ratio for the attention probabilities. TYPE: `float`, optional, defaults to 0.0 DEFAULT: `0.0`
`classifier_dropout`	Argument used when doing token classification, used in the model [`GPTNeoForTokenClassification`]. The dropout ratio for the hidden layer. TYPE: `float`, optional, defaults to 0.1 DEFAULT: `0.1`
`layer_norm_epsilon`	The epsilon used by the layer normalization layers. TYPE: `float`, optional, defaults to 1e-05 DEFAULT: `1e-05`
`initializer_range`	The standard deviation of the truncated_normal_initializer for initializing all weight matrices. TYPE: `float`, optional, defaults to 0.02 DEFAULT: `0.02`
`use_cache`	Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if `config.is_decoder=True`. TYPE: `bool`, optional, defaults to `True` DEFAULT: `True`
`bos_token_id`	The id of the beginning of sentence token in the vocabulary. TYPE: `int`, optional, defaults to 50256 DEFAULT: `50256`
`eos_token_id`	The id of the end of sentence token in the vocabulary. TYPE: `int`, optional, defaults to 50256 DEFAULT: `50256`

>>> from transformers import GPTNeoConfig, GPTNeoModel

>>> # Initializing a GPTNeo EleutherAI/gpt-neo-1.3B style configuration
>>> configuration = GPTNeoConfig()

>>> # Initializing a model (with random weights) from the EleutherAI/gpt-neo-1.3B style configuration
>>> model = GPTNeoModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

Source code in mindnlp\transformers\models\gpt_neo\configuration_gpt_neo.py

class GPTNeoConfig(PretrainedConfig):
    r"""
    This is the configuration class to store the configuration of a [`GPTNeoModel`]. It is used to instantiate a GPT
    Neo 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 GPTNeo
    [EleutherAI/gpt-neo-1.3B](https://huggingface.co/EleutherAI/gpt-neo-1.3B) 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 50257):
            Vocabulary size of the GPT Neo model. Defines the number of different tokens that can be represented by the
            `inputs_ids` passed when calling [`GPTNeoModel`]. Vocabulary size of the model. Defines the different
            tokens that can be represented by the *inputs_ids* passed to the forward method of [`GPTNeoModel`].
        max_position_embeddings (`int`, *optional*, defaults to 2048):
            The maximum sequence length that this model might ever be used with. Typically set this to something large
            just in case (e.g., 512 or 1024 or 2048).
        hidden_size (`int`, *optional*, defaults to 2048):
            Dimensionality of the encoder layers and the pooler layer.
        num_layers (`int`, *optional*, defaults to 24):
            Number of hidden layers in the Transformer encoder.
        attention_types (`List`, *optional*, defaults to `[[['global', 'local'], 12]]`):
            The type of attention for each layer in a `List` of the following format `[[["attention_type"],
            num_layerss]]` e.g. for a 24 layer model `[[["global"], 24]]` or `[[["global", "local"], 12]]` Choose the
            value of `attention_type` from `["global", "local"]`
        num_heads (`int`, *optional*, defaults to 16):
            Number of attention heads for each attention layer in the Transformer encoder.
        intermediate_size (`int`, *optional*, defaults to 8192):
            Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
        window_size (`int`, *optional*, defaults to 256):
            The size of the sliding window for local attention.
        activation_function (`str` or `function`, *optional*, defaults to `"gelu_new"`):
            The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
            `"relu"`, `"selu"` and `"gelu_new"` are supported.
        resid_dropout (`float`, *optional*, defaults to 0.0):
            Residual dropout used in the attention pattern.
        embed_dropout (`float`, *optional*, defaults to 0.0):
            The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
        attention_dropout (`float`, *optional*, defaults to 0.0):
            The dropout ratio for the attention probabilities.
        classifier_dropout (`float`, *optional*, defaults to 0.1):
            Argument used when doing token classification, used in the model [`GPTNeoForTokenClassification`]. The
            dropout ratio for the hidden layer.
        layer_norm_epsilon (`float`, *optional*, defaults to 1e-05):
            The epsilon used by the layer normalization layers.
        initializer_range (`float`, *optional*, defaults to 0.02):
            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
        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`.
        bos_token_id (`int`, *optional*, defaults to 50256):
            The id of the beginning of sentence token in the vocabulary.
        eos_token_id (`int`, *optional*, defaults to 50256):
            The id of the end of sentence token in the vocabulary.

    Example:

    ```python
    >>> from transformers import GPTNeoConfig, GPTNeoModel

    >>> # Initializing a GPTNeo EleutherAI/gpt-neo-1.3B style configuration
    >>> configuration = GPTNeoConfig()

    >>> # Initializing a model (with random weights) from the EleutherAI/gpt-neo-1.3B style configuration
    >>> model = GPTNeoModel(configuration)

    >>> # Accessing the model configuration
    >>> configuration = model.config
    ```"""

    model_type = "gpt_neo"
    keys_to_ignore_at_inference = ["past_key_values"]
    attribute_map = {"num_attention_heads": "num_heads", "num_hidden_layers": "num_layers"}

    def __init__(
        self,
        vocab_size=50257,
        max_position_embeddings=2048,
        hidden_size=2048,
        num_layers=24,
        attention_types=[[["global", "local"], 12]],
        num_heads=16,
        intermediate_size=None,
        window_size=256,
        activation_function="gelu_new",
        resid_dropout=0.0,
        embed_dropout=0.0,
        attention_dropout=0.0,
        classifier_dropout=0.1,
        layer_norm_epsilon=1e-5,
        initializer_range=0.02,
        use_cache=True,
        bos_token_id=50256,
        eos_token_id=50256,
        **kwargs,
    ):
        self.vocab_size = vocab_size
        self.max_position_embeddings = max_position_embeddings
        self.hidden_size = hidden_size
        self.num_layers = num_layers
        self.num_heads = num_heads
        self.intermediate_size = intermediate_size
        self.window_size = window_size
        self.activation_function = activation_function
        self.resid_dropout = resid_dropout
        self.embed_dropout = embed_dropout
        self.attention_dropout = attention_dropout
        self.classifier_dropout = classifier_dropout
        self.layer_norm_epsilon = layer_norm_epsilon
        self.initializer_range = initializer_range
        self.use_cache = use_cache

        self.bos_token_id = bos_token_id
        self.eos_token_id = eos_token_id

        self.attention_types = attention_types
        self.attention_layers = self.expand_attention_types_params(attention_types)

        if len(self.attention_layers) != self.num_layers:
            raise ValueError(
                "Configuration for convolutional module is incorrect. "
                "It is required that `len(config.attention_layers)` == `config.num_layers` "
                f"but is `len(config.attention_layers) = {len(self.attention_layers)}`, "
                f"`config.num_layers = {self.num_layers}`. "
                "`config.attention_layers` is prepared using `config.attention_types`. "
                "Please verify the value of `config.attention_types` argument."
            )

        super().__init__(bos_token_id=bos_token_id, eos_token_id=eos_token_id, **kwargs)

    @staticmethod
    def expand_attention_types_params(attention_types):
        attentions = []
        for item in attention_types:
            for _ in range(item[1]):
                attentions.extend(item[0])
        return attentions