o ei+@s2ddlmZddlmZGdddeZdgZdS))PreTrainedConfig)RopeParametersc:seZdZdZdZdgZdddddddddddd Zd gd gfd d gd gfd gd gfd Z                            d;dedBdedBdedBd edBd!e dBd"e dBd#e dBd$edBd%e dBd&e dBd'edBd(e e e e fBdBd)edBd*edBd+e dBd,e dBd-e dBd.edBd/edBd0e dBd1edBd2edBd3e dBd4e dBd5e dBd6edBd7edBd8edBf8fd9d: ZZS)< DogeConfiga This is the configuration class to store the configuration of a [`DogeModel`]. It is used to instantiate an Doge model according to the specified arguments, defining the model architecture like [SmallDoge/Doge-320M](https://huggingface.co/SmallDoge/Doge-320M). 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 32768): Vocabulary size of the Doge2 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`DogeModel`] hidden_size (`int`, *optional*, defaults to 1024): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 2048): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 32): Number of hidden layers in the Transformer decoder. hidden_dropout (`float`, *optional*, defaults to 0.0): Dropout probability for each sequence transformation and state transformation module. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. 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-06): 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`. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. max_position_embeddings (`int`, *optional*, defaults to 2048): The maximum sequence length that this model might ever be used with. 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`. num_attention_heads (`int`, *optional*, defaults to 8): 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 checkout [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `num_attention_heads`. attention_bias (`bool`, defaults to `False`, *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. sliding_window (`int`, *optional*): Sliding window attention window size. If not specified, will default to `None`. keep_window_size (`int`, *optional*, defaults to 2048): The window size of tokens that are not dynamically masked, and dynamic masking is only performed when the sequence length exceeds this value. is_moe (`bool`, *optional*, defaults to `False`): Whether to use the Cross Domain Mixture of Experts, if `True`, the MoE will inherit the MLP to initialize. num_experts (`int`, *optional*, defaults to 16384): Number of routed experts in the model. This is only used when `is_moe=True`. num_experts_per_tok (`int`, *optional*, defaults to 64): Number of selected experts to route per-token. norm_topk_prob (`bool`, *optional*, defaults to `False`): Whether to normalize the topk probabilities. output_router_logits (`bool`, *optional*, defaults to `False`): Whether or not the router logits should be returned by the model. Enabling this will also allow the model to output the auxiliary loss, including load balancing loss and router z-loss. router_aux_loss_coef (`float`, *optional*, defaults to 0.001): The aux loss factor for the total loss. pad_token_id (`int`, *optional*): Padding token id. bos_token_id (`int`, *optional*): Beginning of stream token id. eos_token_id (`int`, *optional*): End of stream token id. ```python >>> from transformers import DogeConfig, DogeModel >>> # Initializing a Doge-320M style configuration >>> configuration = DogeConfig() >>> # Initializing a model from the Doge-320M style configuration >>> model = DogeModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```dogepast_key_valuescolwiserowwisecolwise_gather_outputrowwise_split_input) zlayers.*.self_attn.q_projzlayers.*.self_attn.k_projzlayers.*.self_attn.v_projzlayers.*.self_attn.dt_projzlayers.*.self_attn.o_projzlayers.*.mlp.gate_projzlayers.*.mlp.up_projzlayers.*.mlp.down_projzlayers.*.mlp.router_gatezlayers.*.mlp.down_embedzlayers.*.mlp.up_embed input_ids inputs_embeds hidden_statesattention_mask) embed_tokenslayersnorm silu{Gz?ư>TFN@@MbP? vocab_size hidden_sizeintermediate_sizenum_hidden_layershidden_dropout hidden_actinitializer_range rms_norm_eps use_cachetie_word_embeddingsmax_position_embeddingsrope_parametersnum_attention_headsnum_key_value_headsattention_biasattention_dropoutmlp_biassliding_windowkeep_window_sizeis_moe num_expertsnum_experts_per_toknorm_topk_proboutput_router_logitsrouter_aux_loss_coef pad_token_id bos_token_id eos_token_idc s||_||_||_||_||_||_||_||_| |_| |_ | |_ ||_ ||_ ||_ ||_||_||_||_||_||_||_||_||_| |_||_||_||_| |_|dur[| |_ tjdi|dS)N)rrr r!r"r#r$r%r&r(r*r+r,r-r.r/r0r1r2r3r4r5r6r'r7r8r9r)super__init__)selfrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2r3r4r5r6r7r8r9kwargs __class__r:i/home/ubuntu/transcripts/venv/lib/python3.10/site-packages/transformers/models/doge/configuration_doge.pyr<s> zDogeConfig.__init__)rrrrrrrrTFrNrNFrFNrFrrFFrNNN)__name__ __module__ __qualname____doc__ model_typekeys_to_ignore_at_inferencebase_model_tp_planbase_model_pp_planintfloatstrboolrdictr< __classcell__r:r:r?rArsW        rN)configuration_utilsrmodeling_rope_utilsrr__all__r:r:r:rAs   5