easydel.modules.xerxes.xerxes_configuration

easydel.modules.xerxes.xerxes_configuration#

class easydel.modules.xerxes.xerxes_configuration.XerxesConfig(vocab_size=256128, hidden_size=4096, intermediate_size=16384, num_hidden_layers=32, num_attention_heads=32, num_key_value_heads=8, head_dim=144, max_position_embeddings=16384, initializer_range=0.02, rms_norm_eps=1e-06, use_cache=True, swish_run=False, pad_token_id=0, eos_token_id=1, bos_token_id=2, num_local_experts: int = 4, xe_moe: bool = True, xe_kvnorm: bool = False, xe_mlpnorm: bool = False, num_experts_per_tok: int = 2, tie_word_embeddings=False, rope_theta=10000.0, rope_scaling=None, layer_types: list[str] | None = None, window_pattern: int | None = None, sliding_window: int | None = None, gradient_checkpointing: EasyDeLGradientCheckPointers = EasyDeLGradientCheckPointers.NONE, bits: int | None = None, scan_layers: bool = False, **kwargs)[source]#

Bases: EasyDeLBaseConfig

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

Parameters

  • vocab_size (int, optional, defaults to 256128) – Vocabulary size of the xerxes model. Defines the number of different tokens that can be represented by the inputs_ids passed to the forward method.

  • hidden_size (int, optional, defaults to 4096) – Dimensionality of the encoder layers and the pooler layer.

  • intermediate_size (int, optional, defaults to 16384) – Dimensionality of the “intermediate” (i.e., feed-forward) layer in the Transformer encoder.

  • num_hidden_layers (int, optional, defaults to 32) – Number of hidden layers in the Transformer encoder.

  • num_attention_heads (int, optional, defaults to 16) – Number of attention heads for each attention layer in the Transformer encoder.

  • num_key_value_heads (int, optional, defaults to 16) – Number of key and value heads for each attention layer in the Transformer encoder.

  • head_dim (int, optional, defaults to 256) – Dimensionality of the attention head.

  • max_position_embeddings (int, optional, defaults to 6144) – The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 2048 or 4096).

  • 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-6) – 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, defaults to 0) – The index of the padding token in the vocabulary.

  • eos_token_id (int, optional, defaults to 1) – The index of the end of sequence token in the vocabulary.

  • bos_token_id (int, optional, defaults to 2) – The index of the beginning of sequence token in the vocabulary.

  • tie_word_embeddings (bool, optional, defaults to True) – Whether to tie the weights of the input embeddings and the output embeddings.

  • rope_theta (float, optional, defaults to 10000.0) – The theta value to use for rotary position embeddings.

  • softmax_scale (float, optional, defaults to 14.9666295471) – softmax scale for attention module.

  • attention_dropout (float, optional, defaults to 0.0) – The dropout ratio for the attention probabilities.

  • gradient_checkpointing (str, optional, defaults to “nothing_saveable”) – The gradient checkpointing configuration.

  • bits (int, optional) – The number of bits to quantize the model to.

  • scan_layers (bool, optional, defaults to False) – Whether to use the scan implementation of the layers.

get_mask_details() → dict[int, easydel.infra.utils.AttnMaskDetail][source]#

Retrieve attention mask details for each layer in the model.

This method generates a dictionary mapping layer indices to their corresponding attention mask details. If a sliding window is defined, each layer is assigned a sliding window attention mask with the specified size.

Returns

A dictionary where keys are layer indices (int) and values are AttnMaskDetail objects specifying the attention mask type and size for each layer.

Return type

dict[int, AttnMaskDetail]

Notes

  • If self.sliding_window is None, an empty dictionary is returned.

  • The method iterates over self.num_hidden_layers to assign mask details for each layer.

  • The attention mask type is set to AttnMaskType.SLIDING when a sliding window is defined.

get_partition_rules(*args, **kwargs)[source]#

Get the partition rules for the Xerxes model (without MoE). :returns: The partition rules. :rtype: tp.Tuple[tp.Tuple[str, PartitionSpec]]

model_type: str = 'xerxes'#
Contents