mindformers.models.bert.BertConfig

class mindformers.models.bert.BertConfig(model_type: str = 'bert', use_one_hot_embeddings: bool = False, num_labels: int = 1, assessment_method: str = '', dropout_prob: float = 0.1, batch_size: int = 16, seq_length: int = 128, vocab_size: int = 30522, hidden_size: int = 768, num_hidden_layers: int = 12, num_attention_heads: int = 12, intermediate_size: int = 3072, hidden_act: str = 'gelu', post_layernorm_residual: bool = True, hidden_dropout_prob: float = 0.1, attention_probs_dropout_prob: float = 0.1, max_position_embeddings: int = 128, type_vocab_size: int = 2, initializer_range: float = 0.02, use_relative_positions: bool = False, dtype: str = 'float32', layernorm_dtype: str = 'float32', softmax_dtype: str = 'float32', compute_dtype: str = 'float16', use_past: bool = False, parallel_config: str = 'default', checkpoint_name_or_path: str = '', moe_config: str = 'default', is_training: bool = True, **kwargs)

BERT config class which defines the model size