{ "architectures": [ "AliceAIForCausalLM" ], "attention_dropout": 0.0, "bos_token_id": 1, "auto_map": { "AutoConfig": "configuration_alice_ai.AliceAIConfig", "AutoModelForCausalLM": "modeling_alice_ai.AliceAIForCausalLM" }, "block_attn_res_block_size": 4, "head_dim": 256, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "kda_allow_negative_eigenvalues": false, "layer_types": [ "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention" ], "eos_token_id": 2, "linear_conv_kernel_dim": 4, "linear_key_head_dim": 128, "linear_num_key_heads": 32, "linear_num_value_heads": 32, "linear_value_head_dim": 128, "max_position_embeddings": 262144, "model_type": "alice_ai", "moe_intermediate_size": 512, "mtp_num_hidden_layers": 1, "num_attention_heads": 16, "number_of_conv_states": 3, "num_experts": 512, "num_experts_per_tok": 10, "num_hidden_layers": 48, "num_key_value_heads": 2, "output_router_logits": false, "partial_rotary_factor": 0.25, "rms_norm_eps": 1e-6, "rope_theta": 1000000.0, "router_bias_correction": true, "router_score_function": "sigmoid", "shared_expert_intermediate_size": 512, "tie_word_embeddings": false, "transformers_version": "5.16.1", "use_cache": true, "vocab_size": 129024 }