{ "architectures": [ "BailingMoeV3ForCausalLM" ], "attention_dropout": 0.0, "auto_map": { "AutoConfig": "configuration_bailing_moe_v3.BailingMoeV3Config", "AutoModel": "modeling_bailing_moe_v3.BailingMoeV3Model", "AutoModelForCausalLM": "modeling_bailing_moe_v3.BailingMoeV3ForCausalLM" }, "dtype": "bfloat16", "embedding_dropout": 0.0, "eos_token_id": 156895, "expert_swiglu_limit_list": [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 4, 4, 4, 4, 4, 4 ], "first_k_dense_replace": 2, "gated_attention_proj_granularity_type": "head_wise", "group_norm_size": 1, "head_dim": 128, "hidden_act": "silu", "hidden_size": 2560, "initializer_range": 0.02, "intermediate_size": 6144, "kda_lower_bound": -5.0, "kda_safe_gate": true, "kv_lora_rank": 512, "layer_group_size": 6, "linear_silu": true, "max_position_embeddings": 131072, "max_window_layers": 20, "moe_intermediate_size": 768, "moe_router_enable_expert_bias": true, "moe_shared_expert_intermediate_size": 768, "mtp_loss_scaling_factor": 0, "mtp_use_kda": false, "n_group": 8, "no_kda_lora": true, "norm_topk_prob": true, "num_attention_heads": 32, "num_experts": 512, "num_experts_per_tok": 8, "num_hidden_layers": 42, "num_key_value_heads": 32, "num_kv_heads_for_linear_attn": 0, "num_nextn_predict_layers": 1, "num_shared_experts": 1, "output_dropout": 0.0, "output_router_logits": false, "pad_token_id": 156892, "partial_rotary_factor": 0.5, "q_lora_rank": null, "qk_head_dim": 192, "qk_nope_head_dim": 128, "qk_rope_head_dim": 64, "quantization_config": { "config_groups": { "group_0": { "format": "nvfp4-pack-quantized", "input_activations": null, "output_activations": null, "targets": [ "Linear" ], "weights": { "actorder": null, "block_structure": null, "dynamic": false, "group_size": 16, "num_bits": 4, "observer": "minmax", "observer_kwargs": {}, "strategy": "tensor_group", "symmetric": true, "type": "float" } } }, "format": "nvfp4-pack-quantized", "global_compression_ratio": null, "ignore": [ "model.layers.0.attention.q_proj", "model.layers.0.attention.k_proj", "model.layers.0.attention.v_proj", "model.layers.0.attention.f_proj", "model.layers.0.attention.b_proj", "model.layers.0.attention.g_proj", "model.layers.0.attention.o_proj", "model.layers.1.attention.q_proj", "model.layers.1.attention.k_proj", "model.layers.1.attention.v_proj", "model.layers.1.attention.f_proj", "model.layers.1.attention.b_proj", "model.layers.1.attention.g_proj", "model.layers.1.attention.o_proj", "model.layers.2.attention.q_proj", "model.layers.2.attention.k_proj", "model.layers.2.attention.v_proj", "model.layers.2.attention.f_proj", "model.layers.2.attention.b_proj", "model.layers.2.attention.g_proj", "model.layers.2.attention.o_proj", "model.layers.2.mlp.shared_experts.gate_proj", "model.layers.2.mlp.shared_experts.up_proj", "model.layers.2.mlp.shared_experts.down_proj", "model.layers.3.attention.q_proj", "model.layers.3.attention.k_proj", "model.layers.3.attention.v_proj", "model.layers.3.attention.f_proj", "model.layers.3.attention.b_proj", "model.layers.3.attention.g_proj", "model.layers.3.attention.o_proj", "model.layers.3.mlp.shared_experts.gate_proj", "model.layers.3.mlp.shared_experts.up_proj", "model.layers.3.mlp.shared_experts.down_proj", "model.layers.4.attention.q_proj", "model.layers.4.attention.k_proj", "model.layers.4.attention.v_proj", "model.layers.4.attention.f_proj", "model.layers.4.attention.b_proj", "model.layers.4.attention.g_proj", "model.layers.4.attention.o_proj", "model.layers.4.mlp.shared_experts.gate_proj", "model.layers.4.mlp.shared_experts.up_proj", "model.layers.4.mlp.shared_experts.down_proj", "model.layers.5.attention.q_proj", "model.layers.5.attention.kv_a_proj_with_mqa", "model.layers.5.attention.kv_b_proj", "model.layers.5.attention.g_proj", "model.layers.5.attention.dense", "model.layers.5.mlp.shared_experts.gate_proj", "model.layers.5.mlp.shared_experts.up_proj", "model.layers.5.mlp.shared_experts.down_proj", "model.layers.6.attention.q_proj", "model.layers.6.attention.k_proj", "model.layers.6.attention.v_proj", "model.layers.6.attention.f_proj", "model.layers.6.attention.b_proj", "model.layers.6.attention.g_proj", "model.layers.6.attention.o_proj", "model.layers.6.mlp.shared_experts.gate_proj", "model.layers.6.mlp.shared_experts.up_proj", "model.layers.6.mlp.shared_experts.down_proj", "model.layers.7.attention.q_proj", "model.layers.7.attention.k_proj", "model.layers.7.attention.v_proj", "model.layers.7.attention.f_proj", "model.layers.7.attention.b_proj", "model.layers.7.attention.g_proj", "model.layers.7.attention.o_proj", "model.layers.7.mlp.shared_experts.gate_proj", "model.layers.7.mlp.shared_experts.up_proj", "model.layers.7.mlp.shared_experts.down_proj", "model.layers.8.attention.q_proj", "model.layers.8.attention.k_proj", "model.layers.8.attention.v_proj", "model.layers.8.attention.f_proj", "model.layers.8.attention.b_proj", "model.layers.8.attention.g_proj", "model.layers.8.attention.o_proj", "model.layers.8.mlp.shared_experts.gate_proj", "model.layers.8.mlp.shared_experts.up_proj", "model.layers.8.mlp.shared_experts.down_proj", "model.layers.9.attention.q_proj", "model.layers.9.attention.k_proj", "model.layers.9.attention.v_proj", "model.layers.9.attention.f_proj", "model.layers.9.attention.b_proj", "model.layers.9.attention.g_proj", "model.layers.9.attention.o_proj", "model.layers.9.mlp.shared_experts.gate_proj", "model.layers.9.mlp.shared_experts.up_proj", "model.layers.9.mlp.shared_experts.down_proj", "model.layers.10.attention.q_proj", "model.layers.10.attention.k_proj", "model.layers.10.attention.v_proj", "model.layers.10.attention.f_proj", "model.layers.10.attention.b_proj", "model.layers.10.attention.g_proj", "model.layers.10.attention.o_proj", "model.layers.10.mlp.shared_experts.gate_proj", "model.layers.10.mlp.shared_experts.up_proj", "model.layers.10.mlp.shared_experts.down_proj", "model.layers.11.attention.q_proj", "model.layers.11.attention.kv_a_proj_with_mqa", "model.layers.11.attention.kv_b_proj", "model.layers.11.attention.g_proj", "model.layers.11.attention.dense", "model.layers.11.mlp.shared_experts.gate_proj", "model.layers.11.mlp.shared_experts.up_proj", "model.layers.11.mlp.shared_experts.down_proj", "model.layers.12.attention.q_proj", "model.layers.12.attention.k_proj", "model.layers.12.attention.v_proj", "model.layers.12.attention.f_proj", "model.layers.12.attention.b_proj", "model.layers.12.attention.g_proj", "model.layers.12.attention.o_proj", "model.layers.12.mlp.shared_experts.gate_proj", "model.layers.12.mlp.shared_experts.up_proj", "model.layers.12.mlp.shared_experts.down_proj", "model.layers.13.attention.q_proj", "model.layers.13.attention.k_proj", "model.layers.13.attention.v_proj", "model.layers.13.attention.f_proj", "model.layers.13.attention.b_proj", "model.layers.13.attention.g_proj", "model.layers.13.attention.o_proj", "model.layers.13.mlp.shared_experts.gate_proj", "model.layers.13.mlp.shared_experts.up_proj", "model.layers.13.mlp.shared_experts.down_proj", "model.layers.14.attention.q_proj", "model.layers.14.attention.k_proj", "model.layers.14.attention.v_proj", "model.layers.14.attention.f_proj", "model.layers.14.attention.b_proj", "model.layers.14.attention.g_proj", "model.layers.14.attention.o_proj", "model.layers.14.mlp.shared_experts.gate_proj", "model.layers.14.mlp.shared_experts.up_proj", "model.layers.14.mlp.shared_experts.down_proj", "model.layers.15.attention.q_proj", "model.layers.15.attention.k_proj", "model.layers.15.attention.v_proj", "model.layers.15.attention.f_proj", "model.layers.15.attention.b_proj", "model.layers.15.attention.g_proj", "model.layers.15.attention.o_proj", "model.layers.15.mlp.shared_experts.gate_proj", "model.layers.15.mlp.shared_experts.up_proj", "model.layers.15.mlp.shared_experts.down_proj", "model.layers.16.attention.q_proj", "model.layers.16.attention.k_proj", "model.layers.16.attention.v_proj", "model.layers.16.attention.f_proj", "model.layers.16.attention.b_proj", "model.layers.16.attention.g_proj", "model.layers.16.attention.o_proj", "model.layers.16.mlp.shared_experts.gate_proj", "model.layers.16.mlp.shared_experts.up_proj", "model.layers.16.mlp.shared_experts.down_proj", "model.layers.17.attention.q_proj", "model.layers.17.attention.kv_a_proj_with_mqa", "model.layers.17.attention.kv_b_proj", "model.layers.17.attention.g_proj", "model.layers.17.attention.dense", "model.layers.17.mlp.shared_experts.gate_proj", "model.layers.17.mlp.shared_experts.up_proj", "model.layers.17.mlp.shared_experts.down_proj", "model.layers.18.attention.q_proj", "model.layers.18.attention.k_proj", "model.layers.18.attention.v_proj", "model.layers.18.attention.f_proj", "model.layers.18.attention.b_proj", "model.layers.18.attention.g_proj", "model.layers.18.attention.o_proj", "model.layers.18.mlp.shared_experts.gate_proj", "model.layers.18.mlp.shared_experts.up_proj", "model.layers.18.mlp.shared_experts.down_proj", "model.layers.19.attention.q_proj", "model.layers.19.attention.k_proj", "model.layers.19.attention.v_proj", "model.layers.19.attention.f_proj", "model.layers.19.attention.b_proj", "model.layers.19.attention.g_proj", "model.layers.19.attention.o_proj", "model.layers.19.mlp.shared_experts.gate_proj", "model.layers.19.mlp.shared_experts.up_proj", "model.layers.19.mlp.shared_experts.down_proj", "model.layers.20.attention.q_proj", "model.layers.20.attention.k_proj", "model.layers.20.attention.v_proj", "model.layers.20.attention.f_proj", "model.layers.20.attention.b_proj", "model.layers.20.attention.g_proj", "model.layers.20.attention.o_proj", "model.layers.20.mlp.shared_experts.gate_proj", "model.layers.20.mlp.shared_experts.up_proj", "model.layers.20.mlp.shared_experts.down_proj", "model.layers.21.attention.q_proj", "model.layers.21.attention.k_proj", "model.layers.21.attention.v_proj", "model.layers.21.attention.f_proj", "model.layers.21.attention.b_proj", "model.layers.21.attention.g_proj", "model.layers.21.attention.o_proj", "model.layers.21.mlp.shared_experts.gate_proj", "model.layers.21.mlp.shared_experts.up_proj", "model.layers.21.mlp.shared_experts.down_proj", "model.layers.22.attention.q_proj", "model.layers.22.attention.k_proj", "model.layers.22.attention.v_proj", "model.layers.22.attention.f_proj", "model.layers.22.attention.b_proj", "model.layers.22.attention.g_proj", "model.layers.22.attention.o_proj", "model.layers.22.mlp.shared_experts.gate_proj", "model.layers.22.mlp.shared_experts.up_proj", "model.layers.22.mlp.shared_experts.down_proj", "model.layers.23.attention.q_proj", "model.layers.23.attention.kv_a_proj_with_mqa", "model.layers.23.attention.kv_b_proj", "model.layers.23.attention.g_proj", "model.layers.23.attention.dense", "model.layers.23.mlp.shared_experts.gate_proj", "model.layers.23.mlp.shared_experts.up_proj", "model.layers.23.mlp.shared_experts.down_proj", "model.layers.24.attention.q_proj", "model.layers.24.attention.k_proj", "model.layers.24.attention.v_proj", "model.layers.24.attention.f_proj", "model.layers.24.attention.b_proj", "model.layers.24.attention.g_proj", "model.layers.24.attention.o_proj", "model.layers.24.mlp.shared_experts.gate_proj", "model.layers.24.mlp.shared_experts.up_proj", "model.layers.24.mlp.shared_experts.down_proj", "model.layers.25.attention.q_proj", "model.layers.25.attention.k_proj", "model.layers.25.attention.v_proj", "model.layers.25.attention.f_proj", "model.layers.25.attention.b_proj", "model.layers.25.attention.g_proj", "model.layers.25.attention.o_proj", "model.layers.25.mlp.shared_experts.gate_proj", "model.layers.25.mlp.shared_experts.up_proj", "model.layers.25.mlp.shared_experts.down_proj", "model.layers.26.attention.q_proj", "model.layers.26.attention.k_proj", "model.layers.26.attention.v_proj", "model.layers.26.attention.f_proj", "model.layers.26.attention.b_proj", "model.layers.26.attention.g_proj", "model.layers.26.attention.o_proj", "model.layers.26.mlp.shared_experts.gate_proj", "model.layers.26.mlp.shared_experts.up_proj", "model.layers.26.mlp.shared_experts.down_proj", "model.layers.27.attention.q_proj", "model.layers.27.attention.k_proj", "model.layers.27.attention.v_proj", "model.layers.27.attention.f_proj", "model.layers.27.attention.b_proj", "model.layers.27.attention.g_proj", "model.layers.27.attention.o_proj", "model.layers.27.mlp.shared_experts.gate_proj", "model.layers.27.mlp.shared_experts.up_proj", "model.layers.27.mlp.shared_experts.down_proj", "model.layers.28.attention.q_proj", "model.layers.28.attention.k_proj", "model.layers.28.attention.v_proj", "model.layers.28.attention.f_proj", "model.layers.28.attention.b_proj", "model.layers.28.attention.g_proj", "model.layers.28.attention.o_proj", "model.layers.28.mlp.shared_experts.gate_proj", "model.layers.28.mlp.shared_experts.up_proj", "model.layers.28.mlp.shared_experts.down_proj", "model.layers.29.attention.q_proj", "model.layers.29.attention.kv_a_proj_with_mqa", "model.layers.29.attention.kv_b_proj", "model.layers.29.attention.g_proj", "model.layers.29.attention.dense", "model.layers.29.mlp.shared_experts.gate_proj", "model.layers.29.mlp.shared_experts.up_proj", "model.layers.29.mlp.shared_experts.down_proj", "model.layers.30.attention.q_proj", "model.layers.30.attention.k_proj", "model.layers.30.attention.v_proj", "model.layers.30.attention.f_proj", "model.layers.30.attention.b_proj", "model.layers.30.attention.g_proj", "model.layers.30.attention.o_proj", "model.layers.30.mlp.shared_experts.gate_proj", "model.layers.30.mlp.shared_experts.up_proj", "model.layers.30.mlp.shared_experts.down_proj", "model.layers.31.attention.q_proj", "model.layers.31.attention.k_proj", "model.layers.31.attention.v_proj", "model.layers.31.attention.f_proj", "model.layers.31.attention.b_proj", "model.layers.31.attention.g_proj", "model.layers.31.attention.o_proj", "model.layers.31.mlp.shared_experts.gate_proj", "model.layers.31.mlp.shared_experts.up_proj", "model.layers.31.mlp.shared_experts.down_proj", "model.layers.32.attention.q_proj", "model.layers.32.attention.k_proj", "model.layers.32.attention.v_proj", "model.layers.32.attention.f_proj", "model.layers.32.attention.b_proj", "model.layers.32.attention.g_proj", "model.layers.32.attention.o_proj", "model.layers.32.mlp.shared_experts.gate_proj", "model.layers.32.mlp.shared_experts.up_proj", "model.layers.32.mlp.shared_experts.down_proj", "model.layers.33.attention.q_proj", "model.layers.33.attention.k_proj", "model.layers.33.attention.v_proj", "model.layers.33.attention.f_proj", "model.layers.33.attention.b_proj", "model.layers.33.attention.g_proj", "model.layers.33.attention.o_proj", "model.layers.33.mlp.shared_experts.gate_proj", "model.layers.33.mlp.shared_experts.up_proj", "model.layers.33.mlp.shared_experts.down_proj", "model.layers.34.attention.q_proj", "model.layers.34.attention.k_proj", "model.layers.34.attention.v_proj", "model.layers.34.attention.f_proj", "model.layers.34.attention.b_proj", "model.layers.34.attention.g_proj", "model.layers.34.attention.o_proj", "model.layers.34.mlp.shared_experts.gate_proj", "model.layers.34.mlp.shared_experts.up_proj", "model.layers.34.mlp.shared_experts.down_proj", "model.layers.35.attention.q_proj", "model.layers.35.attention.kv_a_proj_with_mqa", "model.layers.35.attention.kv_b_proj", "model.layers.35.attention.g_proj", "model.layers.35.attention.dense", "model.layers.35.mlp.shared_experts.gate_proj", "model.layers.35.mlp.shared_experts.up_proj", "model.layers.35.mlp.shared_experts.down_proj", "model.layers.36.attention.q_proj", "model.layers.36.attention.k_proj", "model.layers.36.attention.v_proj", "model.layers.36.attention.f_proj", "model.layers.36.attention.b_proj", "model.layers.36.attention.g_proj", "model.layers.36.attention.o_proj", "model.layers.36.mlp.shared_experts.gate_proj", "model.layers.36.mlp.shared_experts.up_proj", "model.layers.36.mlp.shared_experts.down_proj", "model.layers.37.attention.q_proj", "model.layers.37.attention.k_proj", "model.layers.37.attention.v_proj", "model.layers.37.attention.f_proj", "model.layers.37.attention.b_proj", "model.layers.37.attention.g_proj", "model.layers.37.attention.o_proj", "model.layers.37.mlp.shared_experts.gate_proj", "model.layers.37.mlp.shared_experts.up_proj", "model.layers.37.mlp.shared_experts.down_proj", "model.layers.38.attention.q_proj", "model.layers.38.attention.k_proj", "model.layers.38.attention.v_proj", "model.layers.38.attention.f_proj", "model.layers.38.attention.b_proj", "model.layers.38.attention.g_proj", "model.layers.38.attention.o_proj", "model.layers.38.mlp.shared_experts.gate_proj", "model.layers.38.mlp.shared_experts.up_proj", "model.layers.38.mlp.shared_experts.down_proj", "model.layers.39.attention.q_proj", "model.layers.39.attention.k_proj", "model.layers.39.attention.v_proj", "model.layers.39.attention.f_proj", "model.layers.39.attention.b_proj", "model.layers.39.attention.g_proj", "model.layers.39.attention.o_proj", "model.layers.39.mlp.shared_experts.gate_proj", "model.layers.39.mlp.shared_experts.up_proj", "model.layers.39.mlp.shared_experts.down_proj", "model.layers.40.attention.q_proj", "model.layers.40.attention.k_proj", "model.layers.40.attention.v_proj", "model.layers.40.attention.f_proj", "model.layers.40.attention.b_proj", "model.layers.40.attention.g_proj", "model.layers.40.attention.o_proj", "model.layers.40.mlp.shared_experts.gate_proj", "model.layers.40.mlp.shared_experts.up_proj", "model.layers.40.mlp.shared_experts.down_proj", "model.layers.41.attention.q_proj", "model.layers.41.attention.kv_a_proj_with_mqa", "model.layers.41.attention.kv_b_proj", "model.layers.41.attention.g_proj", "model.layers.41.attention.dense", "model.layers.41.mlp.shared_experts.gate_proj", "model.layers.41.mlp.shared_experts.up_proj", "model.layers.41.mlp.shared_experts.down_proj", "model.layers.42.attention.q_proj", "model.layers.42.attention.kv_a_proj_with_mqa", "model.layers.42.attention.kv_b_proj", "model.layers.42.attention.g_proj", "model.layers.42.attention.dense", "model.layers.42.mlp.shared_experts.gate_proj", "model.layers.42.mlp.shared_experts.up_proj", "model.layers.42.mlp.shared_experts.down_proj", "lm_head" ], "kv_cache_scheme": null, "quant_method": "compressed-tensors", "quantization_status": "compressed", "sparsity_config": {}, "transform_config": {}, "version": "0.12.2" }, "rms_norm_eps": 1e-06, "rope_interleave": true, "rope_scaling": null, "rope_theta": 6000000, "rotary_dim": 64, "routed_scaling_factor": 2.5, "router_dtype": "fp32", "scale_router_input": false, "score_function": "sigmoid", "scoring_func": "sigmoid", "seq_aux": true, "share_expert_swiglu_limit_list": [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 5, 5, 5, 5, 5, 7, 7 ], "short_conv_kernel_size": 4, "tie_word_embeddings": false, "topk_group": 4, "topk_method": "noaux_tc", "transformers_version": "4.56.2", "up_proj_norm": false, "use_bias": false, "use_cache": true, "use_kda_lora": false, "use_mla_nope": false, "use_nGPT": false, "use_qk_norm": true, "use_qkv_bias": false, "v_head_dim": 128, "value_norm": false, "vocab_size": 157184 }