o
    پi)                     @   s`   U d Z ddlmZmZ ddlmZ ddlmZ ee	Z
i Zeeef ed< G dd deZdS )	zEXAONE model configuration    )AnyDict)PretrainedConfig)logging$EXAONE_PRETRAINED_CONFIG_ARCHIVE_MAPc                       sX   e Zd ZdZdZdgZddiZ										
											d fdd	Z  ZS )ExaoneConfiga>  
    This is the configuration class to store the configuration of a :class:`~transformers.ExaoneModel`. It is used to
    instantiate a EXAONE 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 Exaone

    Configuration objects inherit from :class:`~transformers.PretrainedConfig` and can be used to control the model
    outputs. Read the documentation from :class:`~transformers.PretrainedConfig` for more information.


    Args:
        vocab_size (:obj:`int`, `optional`, defaults to 102400):
            Vocabulary size of the EXAONE model. Defines the number of different tokens that can be represented by the
            :obj:`inputs_ids` passed when calling :class:`~transformers.ExaoneModel`. Vocabulary size of the model.
            Defines the different tokens that can be represented by the `inputs_ids` passed to the forward method of
            :class:`~transformers.EXAONEModel`.
        max_position_embeddings (:obj:`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 (:obj:`int`, `optional`, defaults to 2048):
            Dimensionality of the encoder layers and the pooler layer.
        num_layers (:obj:`int`, `optional`, defaults to 32):
            Number of hidden layers in the Transformer encoder.
        num_attention_heads (:obj:`int`, `optional`, defaults to 32):
            Number of attention heads for each attention layer in the Transformer decoder.
        num_key_value_heads (:obj:`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`.
        intermediate_size (:obj:`int`, `optional`, defaults to `hidden_size * 4`):
            Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
        activation_function (:obj:`str` or :obj:`function`, `optional`, defaults to :obj:`"silu"`):
            The non-linear activation function (function or string) in the decoder.
        rope_theta (:obj:`float`, `optional`, defaults to 10000.0):
            The base period of the RoPE embeddings.
        rope_scaling (:obj:`Dict`, `optional`):
            Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type
            and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value
            accordingly.
            Expected contents:
                `rope_type` (:obj:`str`):
                    The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope',
                    'llama3'], with 'default' being the original RoPE implementation.
                `factor` (:obj:`float`, `optional`):
                    Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In
                    most scaling types, a `factor` of x will enable the model to handle sequences of length x *
                    original maximum pre-trained length.
                `original_max_position_embeddings` (:obj:`int`, `optional`):
                    Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during
                    pretraining.
                `attention_factor` (:obj:`float`, `optional`):
                    Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
                    computation. If unspecified, it defaults to value recommended by the implementation, using the
                    `factor` field to infer the suggested value.
                `beta_fast` (:obj:`float`, `optional`):
                    Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
                    ramp function. If unspecified, it defaults to 32.
                `beta_slow` (:obj:`float`, `optional`):
                    Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
                    ramp function. If unspecified, it defaults to 1.
                `short_factor` (:obj:`List[float]`, `optional`):
                    Only used with 'longrope'. The scaling factor to be applied to short contexts (<
                    `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
                    size divided by the number of attention heads divided by 2
                `long_factor` (:obj:`List[float]`, `optional`):
                    Only used with 'longrope'. The scaling factor to be applied to long contexts (<
                    `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden
                    size divided by the number of attention heads divided by 2
                `low_freq_factor` (:obj:`float`, `optional`):
                    Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
                `high_freq_factor` (:obj:`float`, `optional`):
                    Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
        embed_dropout (:obj:`float`, `optional`, defaults to 0.0):
            The dropout probabilitiy for all fully connected layers in the embeddings, encoder, and pooler.
        attention_dropout (:obj:`float`, `optional`, defaults to 0.0):
            The dropout ratio for the attention probabilities.
        layer_norm_epsilon (:obj:`float`, `optional`, defaults to 1e-5):
            The epsilon used by the layer normalization layers.
        initializer_range (:obj:`float`, `optional`, defaults to 0.02):
            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
        use_cache (:obj:`bool`, `optional`, defaults to :obj:`True`):
            Whether or not the model should return the last key/values attentions (not used by all models). Only
            relevant if ``configs.is_decoder=True``.
        bos_token_id (:obj:`int`, `optional`, defaults to 0):
            Beginning of stream token id.
        eos_token_id (:obj:`int`, `optional`, defaults to 2):
            End of stream token id.
        tie_word_embeddings (:obj:`bool`, `optional`, defaults to :obj:`True`):
            Whether to tie weight embeddings
        gradient_checkpointing (:obj:`bool`, `optional`, defaults to :obj:`False`):
            If True, use gradient checkpointing to save memory at the expense of slower backward pass.

        Example::

            >>> from transformers import EXAONEModel, ExaoneConfig

            >>> # Initializing a EXAONE configuration
            >>> configuration = ExaoneConfig()

            >>> # Initializing a model from configuration
            >>> model = EXAONEModel(configuration)

            >>> # Accessing the model configuration
            >>> configuration = model.configs
    exaonepast_key_valuesnum_hidden_layers
num_layers         Nsilu     @        h㈵>{Gz?Tr      c                    s   || _ || _|| _|| _|| _|| _|d u r|}|| _|r!|| _n|d | _|| _|| _	|| _
|| _|| _|| _|	| _|
| _|| _|| _t jd|||d| d S )N   )bos_token_ideos_token_idtie_word_embeddings )
vocab_sizemax_position_embeddingshidden_sizer   num_attention_headsr
   num_key_value_headsintermediate_sizeactivation_functionembed_dropoutattention_dropoutlayer_norm_epsiloninitializer_range	use_cache
rope_thetarope_scalingr   r   super__init__)selfr   r   r   r   r   r   r   r    r&   r'   r!   r"   r#   r$   r%   r   r   r   kwargs	__class__r   M/home/ubuntu/.local/lib/python3.10/site-packages/sglang/srt/configs/exaone.pyr)      s:   

zExaoneConfig.__init__)r   r   r   r   r   NNr   r   Nr   r   r   r   Tr   r   T)	__name__
__module____qualname____doc__
model_typekeys_to_ignore_at_inferenceattribute_mapr)   __classcell__r   r   r,   r.   r      s0    mr   N)r2   typingr   r    transformers.configuration_utilsr   transformers.utilsr   
get_loggerr/   loggerr   str__annotations__r   r   r   r   r.   <module>   s   
