katuni4ka/tiny-random-chatglm2

1from transformers import PretrainedConfig2 3 4class ChatGLMConfig(PretrainedConfig):5    model_type = "chatglm"6    def __init__(7        self,8        num_layers=28,9        padded_vocab_size=65024,10        hidden_size=4096,11        ffn_hidden_size=13696,12        kv_channels=128,13        num_attention_heads=32,14        seq_length=2048,15        hidden_dropout=0.0,16        classifier_dropout=None,17        attention_dropout=0.0,18        layernorm_epsilon=1e-5,19        rmsnorm=True,20        apply_residual_connection_post_layernorm=False,21        post_layer_norm=True,22        add_bias_linear=False,23        add_qkv_bias=False,24        bias_dropout_fusion=True,25        multi_query_attention=False,26        multi_query_group_num=1,27        apply_query_key_layer_scaling=True,28        attention_softmax_in_fp32=True,29        fp32_residual_connection=False,30        quantization_bit=0,31        pre_seq_len=None,32        prefix_projection=False,33        **kwargs34    ):35        self.num_layers = num_layers36        self.vocab_size = padded_vocab_size37        self.padded_vocab_size = padded_vocab_size38        self.hidden_size = hidden_size39        self.ffn_hidden_size = ffn_hidden_size40        self.kv_channels = kv_channels41        self.num_attention_heads = num_attention_heads42        self.seq_length = seq_length43        self.hidden_dropout = hidden_dropout44        self.classifier_dropout = classifier_dropout45        self.attention_dropout = attention_dropout46        self.layernorm_epsilon = layernorm_epsilon47        self.rmsnorm = rmsnorm48        self.apply_residual_connection_post_layernorm = apply_residual_connection_post_layernorm49        self.post_layer_norm = post_layer_norm50        self.add_bias_linear = add_bias_linear51        self.add_qkv_bias = add_qkv_bias52        self.bias_dropout_fusion = bias_dropout_fusion53        self.multi_query_attention = multi_query_attention54        self.multi_query_group_num = multi_query_group_num55        self.apply_query_key_layer_scaling = apply_query_key_layer_scaling56        self.attention_softmax_in_fp32 = attention_softmax_in_fp3257        self.fp32_residual_connection = fp32_residual_connection58        self.quantization_bit = quantization_bit59        self.pre_seq_len = pre_seq_len60        self.prefix_projection = prefix_projection61        super().__init__(**kwargs)