configuration_sl_model.py

from transformers.utils import logging
from transformers.models.llama import LlamaConfig


logger = logging.get_logger(__name__)


class SLModelConfig(LlamaConfig):
    r"""
    This is the configuration class to store the configuration of a [`SLModelModel`]. It is used to instantiate an SLModel
    model according to the specified arguments, defining the model 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 128256):
            Vocabulary size of the SLModel model. Defines the number of different tokens that can be represented by the
            `inputs_ids` passed when calling [`SLModel`]
        hidden_size (`int`, *optional*, defaults to 768):
            Dimensionality of the encoder layers and the pooler layer.
        intermediate_size (`int`, *optional*, defaults to 3072):
            Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder.
        num_hidden_layers (`int`, *optional*, defaults to 12):
            Number of hidden layers in the Transformer encoder.
        num_attention_heads (`int`, *optional*, defaults to 12):
            Number of attention heads for each attention layer in the Transformer encoder.
        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`.
        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
            The non-linear activation function (function or string) in the encoder and pooler.
        max_position_embeddings (`int`, *optional*, defaults to 8192):
            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-05):
            The epsilon used by the rms normalization layers.
        bos_token_id (`int`, *optional*, defaults to 128000):
            Beginning of stream token id.
        eos_token_id (`int`, *optional*, defaults to 128001):
            End of stream token id.
        pad_token_id (`int`, *optional*, defaults to 128001):
            Padding token id.
        mask_token_id (`int`, *optional*, defaults to 128002):
            Mask token id.
        pretraining_tp (`int`, *optional*, defaults to 1):
            Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
            document](https://huggingface.co/docs/transformers/main/perf_train_gpu_many#tensor-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 250000.0):
            The base period of the RoPE embeddings.
        rope_scaling (`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` (`str`):
                    The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope'], 
                    with 'default' being the original RoPE implementation.
                `factor` (`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` (`int`, *optional*):
                    Used with 'dynamic', 'longrope'. The original max position embeddings used during
                    pretraining.
                `attention_factor` (`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` (`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` (`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` (`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` (`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
        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 layers.
        head_dim (`int`, *optional*):
            The attention head dimension. If None, it will default to hidden_size // num_attention_heads
        classifier_pooling (`str`, *optional*, defaults to `"late"`):
            The pooling strategy to use for the classifier. Can be one of ['mean', 'eos'].
        retrieval_pooling (`str`, *optional*, defaults to `"late"`):
            The pooling strategy to use for the retriever. Can be one of ['mean', 'eos'].
    """

    model_type = "sl_model"

    def __init__(
        self,
        vocab_size=128256,
        hidden_size=768,
        intermediate_size=3072,
        num_hidden_layers=12,
        num_attention_heads=12,
        num_key_value_heads=None,
        hidden_act="silu",
        max_position_embeddings=8192,
        initializer_range=0.02,
        rms_norm_eps=1e-05,
        bos_token_id=128000,
        eos_token_id=128001,
        pad_token_id=128001,
        mask_token_id=128002,
        pretraining_tp=1,
        tie_word_embeddings=False,
        rope_theta=250000.0,
        rope_scaling=None,
        attention_bias=False,
        attention_dropout=0.0,
        mlp_bias=False,
        head_dim=None,
        classifier_pooling="mean",
        retrieval_pooling="mean",
        is_causal=False,
        **kwargs,
    ):
        if num_key_value_heads is None:
            num_key_value_heads = num_attention_heads

        if "use_cache" in kwargs:
            kwargs.pop("use_cache", None)

        super().__init__(
            vocab_size=vocab_size,
            hidden_size=hidden_size,
            intermediate_size=intermediate_size,
            num_hidden_layers=num_hidden_layers,
            num_attention_heads=num_attention_heads,
            num_key_value_heads=num_key_value_heads,
            hidden_act=hidden_act,
            max_position_embeddings=max_position_embeddings,
            initializer_range=initializer_range,
            rms_norm_eps=rms_norm_eps,
            use_cache=False,
            bos_token_id=bos_token_id,
            eos_token_id=eos_token_id,
            pad_token_id=pad_token_id,
            pretraining_tp=pretraining_tp,
            tie_word_embeddings=tie_word_embeddings,
            rope_theta=rope_theta,
            rope_scaling=rope_scaling,
            attention_bias=attention_bias,
            attention_dropout=attention_dropout,
            mlp_bias=mlp_bias,
            head_dim=head_dim,
            **kwargs,
        )
        self.mask_token_id = mask_token_id
        self.classifier_pooling = classifier_pooling
        self.retrieval_pooling = retrieval_pooling
        self.is_causal = is_causal


__all__ = ["SLModelConfig"]