{
  "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"
  },
  "embedding_dropout": 0.0,
  "eos_token_id": 156895,
  "expert_swiglu_limit_list": null,
  "first_k_dense_replace": 1,
  "gated_attention_proj_granularity_type": "head_wise",
  "group_norm_size": 1,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 1536,
  "initializer_range": 0.02,
  "intermediate_size": 4608,
  "kda_lower_bound": -5,
  "kda_safe_gate": true,
  "kv_lora_rank": 512,
  "layer_group_size": 4,
  "linear_silu": true,
  "max_position_embeddings": 131072,
  "max_window_layers": 20,
  "moe_intermediate_size": 512,
  "moe_router_enable_expert_bias": true,
  "moe_shared_expert_intermediate_size": 512,
  "mtp_loss_scaling_factor": 0,
  "mtp_use_kda": false,
  "n_group": 8,
  "no_kda_lora": true,
  "norm_topk_prob": true,
  "num_attention_heads": 16,
  "num_experts": 128,
  "num_experts_per_tok": 8,
  "num_hidden_layers": 24,
  "num_key_value_heads": 16,
  "num_kv_heads_for_linear_attn": 0,
  "num_nextn_predict_layers": 0,
  "num_shared_experts": 1,
  "output_dropout": 0.0,
  "output_router_logits": false,
  "pad_token_id": 156892,
  "partial_rotary_factor": 0.5,
  "q_lora_rank": 256,
  "qk_head_dim": 192,
  "qk_nope_head_dim": 128,
  "qk_rope_head_dim": 64,
  "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": null,
  "short_conv_kernel_size": 4,
  "tie_word_embeddings": false,
  "topk_group": 4,
  "topk_method": "noaux_tc",
  "transformers_version": "4.45.0",
  "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,
  "model_type": "bailing_hybrid",
  "torch_dtype": "bfloat16",
  "quantization": {
    "group_size": 64,
    "bits": 4,
    "mode": "affine",
    "model.layers.0.mlp.down_proj": {
      "bits": 5,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.f_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.k_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.o_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.q_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.v_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.2.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.2.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.2.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.f_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.k_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.o_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.q_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.v_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.f_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.g_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.k_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.o_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.q_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.v_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.word_embeddings": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.attention.b_proj": {
      "bits": 5,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.f_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.g_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.k_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.o_proj": {
      "bits": 5,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.12.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.12.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.12.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.13.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.13.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.13.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.14.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.14.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.14.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.16.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.16.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.16.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.17.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.17.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.17.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.18.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.18.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.18.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.dense": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.g_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.20.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.20.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.20.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.21.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.21.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.21.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.22.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.22.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.22.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.embed_q": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.unembed_out": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    }
  },
  "quantization_config": {
    "group_size": 64,
    "bits": 4,
    "mode": "affine",
    "model.layers.0.mlp.down_proj": {
      "bits": 5,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.f_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.k_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.o_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.q_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.attention.v_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.2.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.2.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.2.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.f_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.k_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.o_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.q_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.attention.v_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.f_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.g_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.k_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.o_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.q_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.attention.v_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.word_embeddings": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.attention.b_proj": {
      "bits": 5,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.6.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.f_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.g_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.k_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.attention.o_proj": {
      "bits": 5,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.attention.b_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.9.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.12.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.12.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.12.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.13.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.13.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.13.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.14.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.14.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.14.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.16.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.16.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.16.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.17.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.17.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.17.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.18.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.18.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.18.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.dense": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.g_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.20.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.20.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.20.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.21.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.21.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.21.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.22.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.22.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.22.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.dense": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.g_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.kv_a_proj_with_mqa": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.q_a_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.q_b_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.shared_experts.down_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.shared_experts.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.shared_experts.up_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.1.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.3.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.4.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.5.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.7.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.8.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.10.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.11.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.15.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.mlp.gate.gate_proj": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.embed_q": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.19.attention.unembed_out": {
      "bits": 6,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.mlp.gate.gate_proj": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.embed_q": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    },
    "model.layers.23.attention.unembed_out": {
      "bits": 8,
      "group_size": 64,
      "mode": "affine"
    }
  }
}