{ "vocab_size": 128256, "max_position_embeddings": 131072, "hidden_size": 7168, "intermediate_size": 18432, "moe_intermediate_size": 2048, "num_hidden_layers": 64, "num_nextn_predict_layers": 1, "num_attention_heads": 64, "n_shared_experts": 1, "n_routed_experts": 256, "ep_size": 1, "routed_scaling_factor": 2.5, "kv_lora_rank": 512, "q_lora_rank": 1536, "qk_rope_head_dim": 64, "v_head_dim": 192, "qk_nope_head_dim": 128, "topk_method": "noaux_tc", "n_group": 8, "topk_group": 4, "num_experts_per_tok": 8, "moe_layer_freq": 1, "first_k_dense_replace": 3, "norm_topk_prob": true, "scoring_func": "sigmoid", "num_key_value_heads": 64, "hidden_act": "silu", "initializer_range": 0.006, "rms_norm_eps": 1e-06, "use_cache": true, "rope_theta": 100000, "rope_scaling": { "beta_fast": 32, "beta_slow": 1, "factor": 40.0, "mscale": 1.0, "mscale_all_dim": 1.0, "original_max_position_embeddings": 4096, "type": "yarn" }, "attention_bias": false, "attention_dropout": 0.0, "tie_word_embeddings": false, "architectures": [ "DeepseekV3ForCausalLM" ], "bos_token_id": 1, "eos_token_id": 2, "transformers_version": "4.53.2", "model_type": "deepseek_v3", "torch_dtype": "bfloat16" }