from transformers.configuration_utils import PretrainedConfig
from loguru import  logger as logging


class BuddyGPTConfig(PretrainedConfig):
    """ TinyLLM 配置文件
    """
    
    model_type = "buddygpt"
    keys_to_ignore_at_inference = ["past_key_values"]

    def __init__(
        self, 
        vocab_size=151669,
        hidden_size=4096,
        intermediate_size=4096,
        num_hidden_layers=32,
        num_attention_heads=32,
        num_key_value_heads=None,
        hidden_act="silu",
        num_seq_len=2048,
        initializer_range=0.02,
        rms_norm_eps=1e-6,
        use_cache=True,
        pad_token_id=None,
        bos_token_id=None,
        eos_token_id=None,
        tie_word_embeddings=False,
        rope_theta=10000.0,
        attention_dropout=0.0,
        _attn_implementation="sdpa",
        **kwargs
    ):
        self.vocab_size = vocab_size
        self.num_seq_len = num_seq_len
        self.hidden_size = hidden_size
        self.intermediate_size = intermediate_size
        self.num_hidden_layers = num_hidden_layers
        self.num_attention_heads = num_attention_heads
        
        # for backward compatibility
        if num_key_value_heads is None:
            num_key_value_heads = num_attention_heads
            
        self.num_key_value_heads = num_key_value_heads
        self.hidden_act = hidden_act
        self.initializer_range = initializer_range
        self.rms_norm_eps = rms_norm_eps
        self.use_cache = use_cache
        self.rope_theta = rope_theta
        self.attention_dropout = attention_dropout   
        self._attn_implementation = _attn_implementation
            
        super().__init__(
            pad_token_id=pad_token_id,
            bos_token_id=bos_token_id,
            eos_token_id=eos_token_id,
            tie_word_embeddings=tie_word_embeddings,
            **kwargs
        )