qi<dZddlmZddlmZGddeZdgZy)zDiffLlama model configuration)PreTrainedConfig)RopeParametersc*0eZdZdZdZdgZ ddedzdedzdedzdedzd edzd edzd edzd edzd edzdedzde dzdedzdedzdedzde dzde e ee fzdzde dzdedzdedzdedzf(fd Z xZ S)DiffLlamaConfiga This is the configuration class to store the configuration of a [`DiffLlamaModel`]. It is used to instantiate an DiffLlama 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 [kajuma/DiffLlama-0.3B-handcut](https://huggingface.co/kajuma/DiffLlama-0.3B-handcut). 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 32000): Vocabulary size of the DiffLlama model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`DiffLlamaModel`] hidden_size (`int`, *optional*, defaults to 2048): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 8192): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 16): 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, check out [this paper](https://huggingface.co/papers/2305.13245). 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 decoder. max_position_embeddings (`int`, *optional*, defaults to 2048): 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. 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*): Padding token id. bos_token_id (`int`, *optional*, defaults to 1): Beginning of stream token id. eos_token_id (`int`, *optional*, defaults to 2): End of stream token id. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether to tie weight embeddings rope_parameters (`RopeParameters`, *optional*): Dictionary containing the configuration parameters for the RoPE embeddings. The dictionary should contain a value for `rope_theta` and optionally parameters used for scaling in case you want to use RoPE with longer `max_position_embeddings`. 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. lambda_std_dev (`float`, *optional*, defaults to 0.1): The standard deviation for initialization of parameter lambda in attention layer. head_dim (`int`, *optional*): The attention head dimension. If None, it will default to hidden_size // num_heads ```python >>> from transformers import DiffLlamaModel, DiffLlamaConfig >>> # Initializing a DiffLlama diffllama-7b style configuration >>> configuration = DiffLlamaConfig() >>> # Initializing a model from the diffllama-7b style configuration >>> model = DiffLlamaModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` diffllamapast_key_valuesN vocab_size hidden_sizeintermediate_sizenum_hidden_layersnum_attention_headsnum_key_value_heads hidden_actmax_position_embeddingsinitializer_range rms_norm_eps use_cache pad_token_id bos_token_id eos_token_idtie_word_embeddingsrope_parametersattention_biasattention_dropoutlambda_std_devhead_dimc z||_||_||_||_||_||_||}||_||_| |_| |_ | |_ ||_ ||_ ||_ ||n|j|j z|_||_||_| |_| |_||_t)|Tdi|y)N)r rr r r r rrrrrrrrrrrrrrsuper__init__)selfr r r r r rrrrrrrrrrrrrrrkwargs __class__s g/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/diffllama/configuration_diffllama.pyr zDiffLlamaConfig.__init__es0%'>$&!2!2#6   &"5 #6 $!2(",!2,$,$8d>N>NRVRjRj>j .#6 ((( "6")i}i  Nsilur&g{Gz?gh㈵>TNFNFgg?N)__name__ __module__ __qualname____doc__ model_typekeys_to_ignore_at_inferenceintstrfloatboolrdictr __classcell__)r#s@r$rrsH TJ#4"5"'"&(,(**,*.!'.2*.#'!%#'#$#$+0MQ&+*-'*#+2#$J2#4Z2#: 2# : 2# !4Z 2#!4Z2#$J2#"%t2#!4<2#Dj2#$;2#Dj2#Dj2#Dj2# "D[!2#"($sN/B*CCdJ#2#$t %2#&!4<'2#( )2#**+2#2#r%rN)r/configuration_utilsrmodeling_rope_utilsrr__all__rr%r$r;s*"$31@#&@#F  r%