# cognitivess_model/configuration_cognitivess.py

from transformers import PretrainedConfig

class CognitivessConfig(PretrainedConfig):
    model_type = "cognitivess"

    def __init__(self, hidden_size=4096, num_hidden_layers=32, num_attention_heads=32, intermediate_size=14336,
                 hidden_act="silu", layer_norm_eps=1e-05, max_position_embeddings=8192, vocab_size=128256,
                 bos_token_id=128000, eos_token_id=128001, pad_token_id=0, attention_dropout=0.0, 
                 attention_bias=False, tie_word_embeddings=False, mlp_bias=False, pretraining_tp=1, 
                 rope_scaling=None, rope_theta=500000.0, num_key_value_heads=8, use_cache=True, **kwargs):
        super().__init__(**kwargs)
        self.hidden_size = hidden_size
        self.num_hidden_layers = num_hidden_layers
        self.num_attention_heads = num_attention_heads
        self.intermediate_size = intermediate_size
        self.hidden_act = hidden_act
        self.layer_norm_eps = layer_norm_eps
        self.max_position_embeddings = max_position_embeddings
        self.vocab_size = vocab_size
        self.bos_token_id = bos_token_id
        self.eos_token_id = eos_token_id
        self.pad_token_id = pad_token_id
        self.attention_dropout = attention_dropout
        self.attention_bias = attention_bias
        self.tie_word_embeddings = tie_word_embeddings
        self.mlp_bias = mlp_bias
        self.pretraining_tp = pretraining_tp
        self.rope_scaling = rope_scaling
        self.rope_theta = rope_theta
        self.num_key_value_heads = num_key_value_heads
        self.use_cache = use_cache