{ "architectures": [ "Glm5NextForConditionalGeneration" ], "image_end_token_id": 154831, "image_start_token_id": 154830, "image_token_id": 154854, "model_type": "glm5_next", "quantization_config": { "algo_config": null, "exclude": [ "lm_head", "model.layers.0.mlp.down_proj", "model.layers.0.mlp.gate_proj", "model.layers.0.mlp.up_proj", "model.layers.0.self_attn.b_proj", "model.layers.0.self_attn.f_a_proj", "model.layers.0.self_attn.f_b_proj", "model.layers.0.self_attn.g_a_proj", "model.layers.0.self_attn.g_b_proj", "model.layers.0.self_attn.k_conv1d", "model.layers.0.self_attn.k_proj", "model.layers.0.self_attn.o_proj", "model.layers.0.self_attn.q_conv1d", "model.layers.0.self_attn.q_proj", "model.layers.0.self_attn.v_conv1d", "model.layers.0.self_attn.v_proj", "model.layers.1.mlp.down_proj", "model.layers.1.mlp.gate_proj", "model.layers.1.mlp.up_proj", "model.layers.1.self_attn.b_proj", "model.layers.1.self_attn.f_a_proj", "model.layers.1.self_attn.f_b_proj", "model.layers.1.self_attn.g_a_proj", "model.layers.1.self_attn.g_b_proj", "model.layers.1.self_attn.k_conv1d", "model.layers.1.self_attn.k_proj", "model.layers.1.self_attn.o_proj", "model.layers.1.self_attn.q_conv1d", "model.layers.1.self_attn.q_proj", "model.layers.1.self_attn.v_conv1d", "model.layers.1.self_attn.v_proj", "model.layers.10.mlp.gate", "model.layers.10.mlp.shared_experts.down_proj", "model.layers.10.mlp.shared_experts.gate_proj", "model.layers.10.mlp.shared_experts.up_proj", "model.layers.10.self_attn.b_proj", "model.layers.10.self_attn.f_a_proj", "model.layers.10.self_attn.f_b_proj", "model.layers.10.self_attn.g_a_proj", "model.layers.10.self_attn.g_b_proj", "model.layers.10.self_attn.k_conv1d", "model.layers.10.self_attn.k_proj", "model.layers.10.self_attn.o_proj", "model.layers.10.self_attn.q_conv1d", "model.layers.10.self_attn.q_proj", "model.layers.10.self_attn.v_conv1d", "model.layers.10.self_attn.v_proj", "model.layers.11.mlp.gate", "model.layers.11.mlp.shared_experts.down_proj", "model.layers.11.mlp.shared_experts.gate_proj", "model.layers.11.mlp.shared_experts.up_proj", "model.layers.11.self_attn.indexer.weights_proj", "model.layers.11.self_attn.indexer.wk", "model.layers.11.self_attn.indexer.wq_b", "model.layers.11.self_attn.kv_a_proj_with_mqa", "model.layers.11.self_attn.kv_b_proj", "model.layers.11.self_attn.o_proj", "model.layers.11.self_attn.q_a_proj", "model.layers.11.self_attn.q_b_proj", "model.layers.12.mlp.gate", "model.layers.12.mlp.shared_experts.down_proj", "model.layers.12.mlp.shared_experts.gate_proj", "model.layers.12.mlp.shared_experts.up_proj", "model.layers.12.self_attn.b_proj", "model.layers.12.self_attn.f_a_proj", "model.layers.12.self_attn.f_b_proj", "model.layers.12.self_attn.g_a_proj", "model.layers.12.self_attn.g_b_proj", "model.layers.12.self_attn.k_conv1d", "model.layers.12.self_attn.k_proj", "model.layers.12.self_attn.o_proj", "model.layers.12.self_attn.q_conv1d", "model.layers.12.self_attn.q_proj", "model.layers.12.self_attn.v_conv1d", "model.layers.12.self_attn.v_proj", "model.layers.13.mlp.gate", "model.layers.13.mlp.shared_experts.down_proj", "model.layers.13.mlp.shared_experts.gate_proj", "model.layers.13.mlp.shared_experts.up_proj", "model.layers.13.self_attn.b_proj", "model.layers.13.self_attn.f_a_proj", "model.layers.13.self_attn.f_b_proj", "model.layers.13.self_attn.g_a_proj", "model.layers.13.self_attn.g_b_proj", "model.layers.13.self_attn.k_conv1d", "model.layers.13.self_attn.k_proj", "model.layers.13.self_attn.o_proj", "model.layers.13.self_attn.q_conv1d", "model.layers.13.self_attn.q_proj", "model.layers.13.self_attn.v_conv1d", "model.layers.13.self_attn.v_proj", "model.layers.14.mlp.gate", "model.layers.14.mlp.shared_experts.down_proj", "model.layers.14.mlp.shared_experts.gate_proj", "model.layers.14.mlp.shared_experts.up_proj", "model.layers.14.self_attn.b_proj", "model.layers.14.self_attn.f_a_proj", "model.layers.14.self_attn.f_b_proj", "model.layers.14.self_attn.g_a_proj", "model.layers.14.self_attn.g_b_proj", "model.layers.14.self_attn.k_conv1d", "model.layers.14.self_attn.k_proj", "model.layers.14.self_attn.o_proj", "model.layers.14.self_attn.q_conv1d", "model.layers.14.self_attn.q_proj", "model.layers.14.self_attn.v_conv1d", "model.layers.14.self_attn.v_proj", "model.layers.15.mlp.gate", "model.layers.15.mlp.shared_experts.down_proj", "model.layers.15.mlp.shared_experts.gate_proj", "model.layers.15.mlp.shared_experts.up_proj", "model.layers.15.self_attn.indexer.weights_proj", "model.layers.15.self_attn.indexer.wk", "model.layers.15.self_attn.indexer.wq_b", "model.layers.15.self_attn.kv_a_proj_with_mqa", "model.layers.15.self_attn.kv_b_proj", "model.layers.15.self_attn.o_proj", "model.layers.15.self_attn.q_a_proj", "model.layers.15.self_attn.q_b_proj", "model.layers.16.mlp.gate", "model.layers.16.mlp.shared_experts.down_proj", "model.layers.16.mlp.shared_experts.gate_proj", "model.layers.16.mlp.shared_experts.up_proj", "model.layers.16.self_attn.b_proj", "model.layers.16.self_attn.f_a_proj", "model.layers.16.self_attn.f_b_proj", "model.layers.16.self_attn.g_a_proj", "model.layers.16.self_attn.g_b_proj", "model.layers.16.self_attn.k_conv1d", "model.layers.16.self_attn.k_proj", "model.layers.16.self_attn.o_proj", "model.layers.16.self_attn.q_conv1d", "model.layers.16.self_attn.q_proj", "model.layers.16.self_attn.v_conv1d", "model.layers.16.self_attn.v_proj", "model.layers.17.mlp.gate", "model.layers.17.mlp.shared_experts.down_proj", "model.layers.17.mlp.shared_experts.gate_proj", "model.layers.17.mlp.shared_experts.up_proj", "model.layers.17.self_attn.b_proj", "model.layers.17.self_attn.f_a_proj", "model.layers.17.self_attn.f_b_proj", "model.layers.17.self_attn.g_a_proj", "model.layers.17.self_attn.g_b_proj", "model.layers.17.self_attn.k_conv1d", "model.layers.17.self_attn.k_proj", "model.layers.17.self_attn.o_proj", "model.layers.17.self_attn.q_conv1d", "model.layers.17.self_attn.q_proj", "model.layers.17.self_attn.v_conv1d", "model.layers.17.self_attn.v_proj", "model.layers.18.mlp.gate", "model.layers.18.mlp.shared_experts.down_proj", "model.layers.18.mlp.shared_experts.gate_proj", "model.layers.18.mlp.shared_experts.up_proj", "model.layers.18.self_attn.b_proj", "model.layers.18.self_attn.f_a_proj", "model.layers.18.self_attn.f_b_proj", "model.layers.18.self_attn.g_a_proj", "model.layers.18.self_attn.g_b_proj", "model.layers.18.self_attn.k_conv1d", "model.layers.18.self_attn.k_proj", "model.layers.18.self_attn.o_proj", "model.layers.18.self_attn.q_conv1d", "model.layers.18.self_attn.q_proj", "model.layers.18.self_attn.v_conv1d", "model.layers.18.self_attn.v_proj", "model.layers.19.mlp.gate", "model.layers.19.mlp.shared_experts.down_proj", "model.layers.19.mlp.shared_experts.gate_proj", "model.layers.19.mlp.shared_experts.up_proj", "model.layers.19.self_attn.indexer.weights_proj", "model.layers.19.self_attn.indexer.wk", "model.layers.19.self_attn.indexer.wq_b", "model.layers.19.self_attn.kv_a_proj_with_mqa", "model.layers.19.self_attn.kv_b_proj", "model.layers.19.self_attn.o_proj", "model.layers.19.self_attn.q_a_proj", "model.layers.19.self_attn.q_b_proj", "model.layers.2.mlp.down_proj", "model.layers.2.mlp.gate_proj", "model.layers.2.mlp.up_proj", "model.layers.2.self_attn.b_proj", "model.layers.2.self_attn.f_a_proj", "model.layers.2.self_attn.f_b_proj", "model.layers.2.self_attn.g_a_proj", "model.layers.2.self_attn.g_b_proj", "model.layers.2.self_attn.k_conv1d", "model.layers.2.self_attn.k_proj", "model.layers.2.self_attn.o_proj", "model.layers.2.self_attn.q_conv1d", "model.layers.2.self_attn.q_proj", "model.layers.2.self_attn.v_conv1d", "model.layers.2.self_attn.v_proj", "model.layers.20.mlp.gate", "model.layers.20.mlp.shared_experts.down_proj", "model.layers.20.mlp.shared_experts.gate_proj", "model.layers.20.mlp.shared_experts.up_proj", "model.layers.20.self_attn.b_proj", "model.layers.20.self_attn.f_a_proj", "model.layers.20.self_attn.f_b_proj", "model.layers.20.self_attn.g_a_proj", "model.layers.20.self_attn.g_b_proj", "model.layers.20.self_attn.k_conv1d", "model.layers.20.self_attn.k_proj", "model.layers.20.self_attn.o_proj", "model.layers.20.self_attn.q_conv1d", "model.layers.20.self_attn.q_proj", "model.layers.20.self_attn.v_conv1d", "model.layers.20.self_attn.v_proj", "model.layers.21.mlp.gate", "model.layers.21.mlp.shared_experts.down_proj", "model.layers.21.mlp.shared_experts.gate_proj", "model.layers.21.mlp.shared_experts.up_proj", "model.layers.21.self_attn.b_proj", "model.layers.21.self_attn.f_a_proj", "model.layers.21.self_attn.f_b_proj", "model.layers.21.self_attn.g_a_proj", "model.layers.21.self_attn.g_b_proj", "model.layers.21.self_attn.k_conv1d", "model.layers.21.self_attn.k_proj", "model.layers.21.self_attn.o_proj", "model.layers.21.self_attn.q_conv1d", "model.layers.21.self_attn.q_proj", "model.layers.21.self_attn.v_conv1d", "model.layers.21.self_attn.v_proj", "model.layers.22.mlp.gate", "model.layers.22.mlp.shared_experts.down_proj", "model.layers.22.mlp.shared_experts.gate_proj", "model.layers.22.mlp.shared_experts.up_proj", "model.layers.22.self_attn.b_proj", "model.layers.22.self_attn.f_a_proj", "model.layers.22.self_attn.f_b_proj", "model.layers.22.self_attn.g_a_proj", "model.layers.22.self_attn.g_b_proj", "model.layers.22.self_attn.k_conv1d", "model.layers.22.self_attn.k_proj", "model.layers.22.self_attn.o_proj", "model.layers.22.self_attn.q_conv1d", "model.layers.22.self_attn.q_proj", "model.layers.22.self_attn.v_conv1d", "model.layers.22.self_attn.v_proj", "model.layers.23.mlp.gate", "model.layers.23.mlp.shared_experts.down_proj", "model.layers.23.mlp.shared_experts.gate_proj", "model.layers.23.mlp.shared_experts.up_proj", "model.layers.23.self_attn.indexer.weights_proj", "model.layers.23.self_attn.indexer.wk", "model.layers.23.self_attn.indexer.wq_b", "model.layers.23.self_attn.kv_a_proj_with_mqa", "model.layers.23.self_attn.kv_b_proj", "model.layers.23.self_attn.o_proj", "model.layers.23.self_attn.q_a_proj", "model.layers.23.self_attn.q_b_proj", "model.layers.24.mlp.gate", "model.layers.24.mlp.shared_experts.down_proj", "model.layers.24.mlp.shared_experts.gate_proj", "model.layers.24.mlp.shared_experts.up_proj", "model.layers.24.self_attn.b_proj", "model.layers.24.self_attn.f_a_proj", "model.layers.24.self_attn.f_b_proj", "model.layers.24.self_attn.g_a_proj", "model.layers.24.self_attn.g_b_proj", "model.layers.24.self_attn.k_conv1d", "model.layers.24.self_attn.k_proj", "model.layers.24.self_attn.o_proj", "model.layers.24.self_attn.q_conv1d", "model.layers.24.self_attn.q_proj", "model.layers.24.self_attn.v_conv1d", "model.layers.24.self_attn.v_proj", "model.layers.25.mlp.gate", "model.layers.25.mlp.shared_experts.down_proj", "model.layers.25.mlp.shared_experts.gate_proj", "model.layers.25.mlp.shared_experts.up_proj", "model.layers.25.self_attn.b_proj", "model.layers.25.self_attn.f_a_proj", "model.layers.25.self_attn.f_b_proj", "model.layers.25.self_attn.g_a_proj", "model.layers.25.self_attn.g_b_proj", "model.layers.25.self_attn.k_conv1d", "model.layers.25.self_attn.k_proj", "model.layers.25.self_attn.o_proj", "model.layers.25.self_attn.q_conv1d", "model.layers.25.self_attn.q_proj", "model.layers.25.self_attn.v_conv1d", "model.layers.25.self_attn.v_proj", "model.layers.26.mlp.gate", "model.layers.26.mlp.shared_experts.down_proj", "model.layers.26.mlp.shared_experts.gate_proj", "model.layers.26.mlp.shared_experts.up_proj", "model.layers.26.self_attn.b_proj", "model.layers.26.self_attn.f_a_proj", "model.layers.26.self_attn.f_b_proj", "model.layers.26.self_attn.g_a_proj", "model.layers.26.self_attn.g_b_proj", "model.layers.26.self_attn.k_conv1d", "model.layers.26.self_attn.k_proj", "model.layers.26.self_attn.o_proj", "model.layers.26.self_attn.q_conv1d", "model.layers.26.self_attn.q_proj", "model.layers.26.self_attn.v_conv1d", "model.layers.26.self_attn.v_proj", "model.layers.27.mlp.gate", "model.layers.27.mlp.shared_experts.down_proj", "model.layers.27.mlp.shared_experts.gate_proj", "model.layers.27.mlp.shared_experts.up_proj", "model.layers.27.self_attn.indexer.weights_proj", "model.layers.27.self_attn.indexer.wk", "model.layers.27.self_attn.indexer.wq_b", "model.layers.27.self_attn.kv_a_proj_with_mqa", "model.layers.27.self_attn.kv_b_proj", "model.layers.27.self_attn.o_proj", "model.layers.27.self_attn.q_a_proj", "model.layers.27.self_attn.q_b_proj", "model.layers.28.mlp.gate", "model.layers.28.mlp.shared_experts.down_proj", "model.layers.28.mlp.shared_experts.gate_proj", "model.layers.28.mlp.shared_experts.up_proj", "model.layers.28.self_attn.b_proj", "model.layers.28.self_attn.f_a_proj", "model.layers.28.self_attn.f_b_proj", "model.layers.28.self_attn.g_a_proj", "model.layers.28.self_attn.g_b_proj", "model.layers.28.self_attn.k_conv1d", "model.layers.28.self_attn.k_proj", "model.layers.28.self_attn.o_proj", "model.layers.28.self_attn.q_conv1d", "model.layers.28.self_attn.q_proj", "model.layers.28.self_attn.v_conv1d", "model.layers.28.self_attn.v_proj", "model.layers.29.mlp.gate", "model.layers.29.mlp.shared_experts.down_proj", "model.layers.29.mlp.shared_experts.gate_proj", "model.layers.29.mlp.shared_experts.up_proj", "model.layers.29.self_attn.b_proj", "model.layers.29.self_attn.f_a_proj", "model.layers.29.self_attn.f_b_proj", "model.layers.29.self_attn.g_a_proj", "model.layers.29.self_attn.g_b_proj", "model.layers.29.self_attn.k_conv1d", "model.layers.29.self_attn.k_proj", "model.layers.29.self_attn.o_proj", "model.layers.29.self_attn.q_conv1d", "model.layers.29.self_attn.q_proj", "model.layers.29.self_attn.v_conv1d", "model.layers.29.self_attn.v_proj", "model.layers.3.mlp.gate", "model.layers.3.mlp.shared_experts.down_proj", "model.layers.3.mlp.shared_experts.gate_proj", "model.layers.3.mlp.shared_experts.up_proj", "model.layers.3.self_attn.indexer.weights_proj", "model.layers.3.self_attn.indexer.wk", "model.layers.3.self_attn.indexer.wq_b", "model.layers.3.self_attn.kv_a_proj_with_mqa", "model.layers.3.self_attn.kv_b_proj", "model.layers.3.self_attn.o_proj", "model.layers.3.self_attn.q_a_proj", "model.layers.3.self_attn.q_b_proj", "model.layers.30.mlp.gate", "model.layers.30.mlp.shared_experts.down_proj", "model.layers.30.mlp.shared_experts.gate_proj", "model.layers.30.mlp.shared_experts.up_proj", "model.layers.30.self_attn.b_proj", "model.layers.30.self_attn.f_a_proj", "model.layers.30.self_attn.f_b_proj", "model.layers.30.self_attn.g_a_proj", "model.layers.30.self_attn.g_b_proj", "model.layers.30.self_attn.k_conv1d", "model.layers.30.self_attn.k_proj", "model.layers.30.self_attn.o_proj", "model.layers.30.self_attn.q_conv1d", "model.layers.30.self_attn.q_proj", "model.layers.30.self_attn.v_conv1d", "model.layers.30.self_attn.v_proj", "model.layers.31.mlp.gate", "model.layers.31.mlp.shared_experts.down_proj", "model.layers.31.mlp.shared_experts.gate_proj", "model.layers.31.mlp.shared_experts.up_proj", "model.layers.31.self_attn.indexer.weights_proj", "model.layers.31.self_attn.indexer.wk", "model.layers.31.self_attn.indexer.wq_b", "model.layers.31.self_attn.kv_a_proj_with_mqa", "model.layers.31.self_attn.kv_b_proj", "model.layers.31.self_attn.o_proj", "model.layers.31.self_attn.q_a_proj", "model.layers.31.self_attn.q_b_proj", "model.layers.32.mlp.gate", "model.layers.32.mlp.shared_experts.down_proj", "model.layers.32.mlp.shared_experts.gate_proj", "model.layers.32.mlp.shared_experts.up_proj", "model.layers.32.self_attn.b_proj", "model.layers.32.self_attn.f_a_proj", "model.layers.32.self_attn.f_b_proj", "model.layers.32.self_attn.g_a_proj", "model.layers.32.self_attn.g_b_proj", "model.layers.32.self_attn.k_conv1d", "model.layers.32.self_attn.k_proj", "model.layers.32.self_attn.o_proj", "model.layers.32.self_attn.q_conv1d", "model.layers.32.self_attn.q_proj", "model.layers.32.self_attn.v_conv1d", "model.layers.32.self_attn.v_proj", "model.layers.33.mlp.gate", "model.layers.33.mlp.shared_experts.down_proj", "model.layers.33.mlp.shared_experts.gate_proj", "model.layers.33.mlp.shared_experts.up_proj", "model.layers.33.self_attn.b_proj", "model.layers.33.self_attn.f_a_proj", "model.layers.33.self_attn.f_b_proj", "model.layers.33.self_attn.g_a_proj", "model.layers.33.self_attn.g_b_proj", "model.layers.33.self_attn.k_conv1d", "model.layers.33.self_attn.k_proj", "model.layers.33.self_attn.o_proj", "model.layers.33.self_attn.q_conv1d", "model.layers.33.self_attn.q_proj", "model.layers.33.self_attn.v_conv1d", "model.layers.33.self_attn.v_proj", "model.layers.34.mlp.gate", "model.layers.34.mlp.shared_experts.down_proj", "model.layers.34.mlp.shared_experts.gate_proj", "model.layers.34.mlp.shared_experts.up_proj", "model.layers.34.self_attn.b_proj", "model.layers.34.self_attn.f_a_proj", "model.layers.34.self_attn.f_b_proj", "model.layers.34.self_attn.g_a_proj", "model.layers.34.self_attn.g_b_proj", "model.layers.34.self_attn.k_conv1d", "model.layers.34.self_attn.k_proj", "model.layers.34.self_attn.o_proj", "model.layers.34.self_attn.q_conv1d", "model.layers.34.self_attn.q_proj", "model.layers.34.self_attn.v_conv1d", "model.layers.34.self_attn.v_proj", "model.layers.35.mlp.gate", "model.layers.35.mlp.shared_experts.down_proj", "model.layers.35.mlp.shared_experts.gate_proj", "model.layers.35.mlp.shared_experts.up_proj", "model.layers.35.self_attn.indexer.weights_proj", "model.layers.35.self_attn.indexer.wk", "model.layers.35.self_attn.indexer.wq_b", "model.layers.35.self_attn.kv_a_proj_with_mqa", "model.layers.35.self_attn.kv_b_proj", "model.layers.35.self_attn.o_proj", "model.layers.35.self_attn.q_a_proj", "model.layers.35.self_attn.q_b_proj", "model.layers.36.mlp.gate", "model.layers.36.mlp.shared_experts.down_proj", "model.layers.36.mlp.shared_experts.gate_proj", "model.layers.36.mlp.shared_experts.up_proj", "model.layers.36.self_attn.b_proj", "model.layers.36.self_attn.f_a_proj", "model.layers.36.self_attn.f_b_proj", "model.layers.36.self_attn.g_a_proj", "model.layers.36.self_attn.g_b_proj", "model.layers.36.self_attn.k_conv1d", "model.layers.36.self_attn.k_proj", "model.layers.36.self_attn.o_proj", "model.layers.36.self_attn.q_conv1d", "model.layers.36.self_attn.q_proj", "model.layers.36.self_attn.v_conv1d", "model.layers.36.self_attn.v_proj", "model.layers.37.mlp.gate", "model.layers.37.mlp.shared_experts.down_proj", "model.layers.37.mlp.shared_experts.gate_proj", "model.layers.37.mlp.shared_experts.up_proj", "model.layers.37.self_attn.b_proj", "model.layers.37.self_attn.f_a_proj", "model.layers.37.self_attn.f_b_proj", "model.layers.37.self_attn.g_a_proj", "model.layers.37.self_attn.g_b_proj", "model.layers.37.self_attn.k_conv1d", "model.layers.37.self_attn.k_proj", "model.layers.37.self_attn.o_proj", "model.layers.37.self_attn.q_conv1d", "model.layers.37.self_attn.q_proj", "model.layers.37.self_attn.v_conv1d", "model.layers.37.self_attn.v_proj", "model.layers.38.mlp.gate", "model.layers.38.mlp.shared_experts.down_proj", "model.layers.38.mlp.shared_experts.gate_proj", "model.layers.38.mlp.shared_experts.up_proj", "model.layers.38.self_attn.b_proj", "model.layers.38.self_attn.f_a_proj", "model.layers.38.self_attn.f_b_proj", "model.layers.38.self_attn.g_a_proj", "model.layers.38.self_attn.g_b_proj", "model.layers.38.self_attn.k_conv1d", "model.layers.38.self_attn.k_proj", "model.layers.38.self_attn.o_proj", "model.layers.38.self_attn.q_conv1d", "model.layers.38.self_attn.q_proj", "model.layers.38.self_attn.v_conv1d", "model.layers.38.self_attn.v_proj", "model.layers.39.mlp.gate", "model.layers.39.mlp.shared_experts.down_proj", "model.layers.39.mlp.shared_experts.gate_proj", "model.layers.39.mlp.shared_experts.up_proj", "model.layers.39.self_attn.indexer.weights_proj", "model.layers.39.self_attn.indexer.wk", "model.layers.39.self_attn.indexer.wq_b", "model.layers.39.self_attn.kv_a_proj_with_mqa", "model.layers.39.self_attn.kv_b_proj", "model.layers.39.self_attn.o_proj", "model.layers.39.self_attn.q_a_proj", "model.layers.39.self_attn.q_b_proj", "model.layers.4.mlp.gate", "model.layers.4.mlp.shared_experts.down_proj", "model.layers.4.mlp.shared_experts.gate_proj", "model.layers.4.mlp.shared_experts.up_proj", "model.layers.4.self_attn.b_proj", "model.layers.4.self_attn.f_a_proj", "model.layers.4.self_attn.f_b_proj", "model.layers.4.self_attn.g_a_proj", "model.layers.4.self_attn.g_b_proj", "model.layers.4.self_attn.k_conv1d", "model.layers.4.self_attn.k_proj", "model.layers.4.self_attn.o_proj", "model.layers.4.self_attn.q_conv1d", "model.layers.4.self_attn.q_proj", "model.layers.4.self_attn.v_conv1d", "model.layers.4.self_attn.v_proj", "model.layers.40.mlp.gate", "model.layers.40.mlp.shared_experts.down_proj", "model.layers.40.mlp.shared_experts.gate_proj", "model.layers.40.mlp.shared_experts.up_proj", "model.layers.40.self_attn.b_proj", "model.layers.40.self_attn.f_a_proj", "model.layers.40.self_attn.f_b_proj", "model.layers.40.self_attn.g_a_proj", "model.layers.40.self_attn.g_b_proj", "model.layers.40.self_attn.k_conv1d", "model.layers.40.self_attn.k_proj", "model.layers.40.self_attn.o_proj", "model.layers.40.self_attn.q_conv1d", "model.layers.40.self_attn.q_proj", "model.layers.40.self_attn.v_conv1d", "model.layers.40.self_attn.v_proj", "model.layers.41.mlp.gate", "model.layers.41.mlp.shared_experts.down_proj", "model.layers.41.mlp.shared_experts.gate_proj", "model.layers.41.mlp.shared_experts.up_proj", "model.layers.41.self_attn.b_proj", "model.layers.41.self_attn.f_a_proj", "model.layers.41.self_attn.f_b_proj", "model.layers.41.self_attn.g_a_proj", "model.layers.41.self_attn.g_b_proj", "model.layers.41.self_attn.k_conv1d", "model.layers.41.self_attn.k_proj", "model.layers.41.self_attn.o_proj", "model.layers.41.self_attn.q_conv1d", "model.layers.41.self_attn.q_proj", "model.layers.41.self_attn.v_conv1d", "model.layers.41.self_attn.v_proj", "model.layers.42.mlp.gate", "model.layers.42.mlp.shared_experts.down_proj", "model.layers.42.mlp.shared_experts.gate_proj", "model.layers.42.mlp.shared_experts.up_proj", "model.layers.42.self_attn.b_proj", "model.layers.42.self_attn.f_a_proj", "model.layers.42.self_attn.f_b_proj", "model.layers.42.self_attn.g_a_proj", "model.layers.42.self_attn.g_b_proj", "model.layers.42.self_attn.k_conv1d", "model.layers.42.self_attn.k_proj", "model.layers.42.self_attn.o_proj", "model.layers.42.self_attn.q_conv1d", "model.layers.42.self_attn.q_proj", "model.layers.42.self_attn.v_conv1d", "model.layers.42.self_attn.v_proj", "model.layers.43.mlp.gate", "model.layers.43.mlp.shared_experts.down_proj", "model.layers.43.mlp.shared_experts.gate_proj", "model.layers.43.mlp.shared_experts.up_proj", "model.layers.43.self_attn.indexer.weights_proj", "model.layers.43.self_attn.indexer.wk", "model.layers.43.self_attn.indexer.wq_b", "model.layers.43.self_attn.kv_a_proj_with_mqa", "model.layers.43.self_attn.kv_b_proj", "model.layers.43.self_attn.o_proj", "model.layers.43.self_attn.q_a_proj", "model.layers.43.self_attn.q_b_proj", "model.layers.44.mlp.gate", "model.layers.44.mlp.shared_experts.down_proj", "model.layers.44.mlp.shared_experts.gate_proj", "model.layers.44.mlp.shared_experts.up_proj", "model.layers.44.self_attn.b_proj", "model.layers.44.self_attn.f_a_proj", "model.layers.44.self_attn.f_b_proj", "model.layers.44.self_attn.g_a_proj", "model.layers.44.self_attn.g_b_proj", "model.layers.44.self_attn.k_conv1d", "model.layers.44.self_attn.k_proj", "model.layers.44.self_attn.o_proj", "model.layers.44.self_attn.q_conv1d", "model.layers.44.self_attn.q_proj", "model.layers.44.self_attn.v_conv1d", "model.layers.44.self_attn.v_proj", "model.layers.45.eh_proj", "model.layers.45.mlp.gate", "model.layers.45.mlp.shared_experts.down_proj", "model.layers.45.mlp.shared_experts.gate_proj", "model.layers.45.mlp.shared_experts.up_proj", "model.layers.45.self_attn.indexer.weights_proj", "model.layers.45.self_attn.indexer.wk", "model.layers.45.self_attn.indexer.wq_b", "model.layers.45.self_attn.kv_a_proj_with_mqa", "model.layers.45.self_attn.kv_b_proj", "model.layers.45.self_attn.o_proj", "model.layers.45.self_attn.q_a_proj", "model.layers.45.self_attn.q_b_proj", "model.layers.5.mlp.gate", "model.layers.5.mlp.shared_experts.down_proj", "model.layers.5.mlp.shared_experts.gate_proj", "model.layers.5.mlp.shared_experts.up_proj", "model.layers.5.self_attn.b_proj", "model.layers.5.self_attn.f_a_proj", "model.layers.5.self_attn.f_b_proj", "model.layers.5.self_attn.g_a_proj", "model.layers.5.self_attn.g_b_proj", "model.layers.5.self_attn.k_conv1d", "model.layers.5.self_attn.k_proj", "model.layers.5.self_attn.o_proj", "model.layers.5.self_attn.q_conv1d", "model.layers.5.self_attn.q_proj", "model.layers.5.self_attn.v_conv1d", "model.layers.5.self_attn.v_proj", "model.layers.6.mlp.gate", "model.layers.6.mlp.shared_experts.down_proj", "model.layers.6.mlp.shared_experts.gate_proj", "model.layers.6.mlp.shared_experts.up_proj", "model.layers.6.self_attn.b_proj", "model.layers.6.self_attn.f_a_proj", "model.layers.6.self_attn.f_b_proj", "model.layers.6.self_attn.g_a_proj", "model.layers.6.self_attn.g_b_proj", "model.layers.6.self_attn.k_conv1d", "model.layers.6.self_attn.k_proj", "model.layers.6.self_attn.o_proj", "model.layers.6.self_attn.q_conv1d", "model.layers.6.self_attn.q_proj", "model.layers.6.self_attn.v_conv1d", "model.layers.6.self_attn.v_proj", "model.layers.7.mlp.gate", "model.layers.7.mlp.shared_experts.down_proj", "model.layers.7.mlp.shared_experts.gate_proj", "model.layers.7.mlp.shared_experts.up_proj", "model.layers.7.self_attn.indexer.weights_proj", "model.layers.7.self_attn.indexer.wk", "model.layers.7.self_attn.indexer.wq_b", "model.layers.7.self_attn.kv_a_proj_with_mqa", "model.layers.7.self_attn.kv_b_proj", "model.layers.7.self_attn.o_proj", "model.layers.7.self_attn.q_a_proj", "model.layers.7.self_attn.q_b_proj", "model.layers.8.mlp.gate", "model.layers.8.mlp.shared_experts.down_proj", "model.layers.8.mlp.shared_experts.gate_proj", "model.layers.8.mlp.shared_experts.up_proj", "model.layers.8.self_attn.b_proj", "model.layers.8.self_attn.f_a_proj", "model.layers.8.self_attn.f_b_proj", "model.layers.8.self_attn.g_a_proj", "model.layers.8.self_attn.g_b_proj", "model.layers.8.self_attn.k_conv1d", "model.layers.8.self_attn.k_proj", "model.layers.8.self_attn.o_proj", "model.layers.8.self_attn.q_conv1d", "model.layers.8.self_attn.q_proj", "model.layers.8.self_attn.v_conv1d", "model.layers.8.self_attn.v_proj", "model.layers.9.mlp.gate", "model.layers.9.mlp.shared_experts.down_proj", "model.layers.9.mlp.shared_experts.gate_proj", "model.layers.9.mlp.shared_experts.up_proj", "model.layers.9.self_attn.b_proj", "model.layers.9.self_attn.f_a_proj", "model.layers.9.self_attn.f_b_proj", "model.layers.9.self_attn.g_a_proj", "model.layers.9.self_attn.g_b_proj", "model.layers.9.self_attn.k_conv1d", "model.layers.9.self_attn.k_proj", "model.layers.9.self_attn.o_proj", "model.layers.9.self_attn.q_conv1d", "model.layers.9.self_attn.q_proj", "model.layers.9.self_attn.v_conv1d", "model.layers.9.self_attn.v_proj", "visual.blocks.0.attn.proj", "visual.blocks.0.attn.qkv_proj", "visual.blocks.0.mlp.down_proj", "visual.blocks.0.mlp.gate_proj", "visual.blocks.0.mlp.up_proj", "visual.blocks.1.attn.proj", "visual.blocks.1.attn.qkv_proj", "visual.blocks.1.mlp.down_proj", "visual.blocks.1.mlp.gate_proj", "visual.blocks.1.mlp.up_proj", "visual.blocks.10.attn.proj", "visual.blocks.10.attn.qkv_proj", "visual.blocks.10.mlp.down_proj", "visual.blocks.10.mlp.gate_proj", "visual.blocks.10.mlp.up_proj", "visual.blocks.11.attn.proj", "visual.blocks.11.attn.qkv_proj", "visual.blocks.11.mlp.down_proj", "visual.blocks.11.mlp.gate_proj", "visual.blocks.11.mlp.up_proj", "visual.blocks.12.attn.proj", "visual.blocks.12.attn.qkv_proj", "visual.blocks.12.mlp.down_proj", "visual.blocks.12.mlp.gate_proj", "visual.blocks.12.mlp.up_proj", "visual.blocks.13.attn.proj", "visual.blocks.13.attn.qkv_proj", "visual.blocks.13.mlp.down_proj", "visual.blocks.13.mlp.gate_proj", "visual.blocks.13.mlp.up_proj", "visual.blocks.14.attn.proj", "visual.blocks.14.attn.qkv_proj", "visual.blocks.14.mlp.down_proj", "visual.blocks.14.mlp.gate_proj", "visual.blocks.14.mlp.up_proj", "visual.blocks.15.attn.proj", "visual.blocks.15.attn.qkv_proj", "visual.blocks.15.mlp.down_proj", "visual.blocks.15.mlp.gate_proj", "visual.blocks.15.mlp.up_proj", "visual.blocks.16.attn.proj", "visual.blocks.16.attn.qkv_proj", "visual.blocks.16.mlp.down_proj", "visual.blocks.16.mlp.gate_proj", "visual.blocks.16.mlp.up_proj", "visual.blocks.17.attn.proj", "visual.blocks.17.attn.qkv_proj", "visual.blocks.17.mlp.down_proj", "visual.blocks.17.mlp.gate_proj", "visual.blocks.17.mlp.up_proj", "visual.blocks.18.attn.proj", "visual.blocks.18.attn.qkv_proj", "visual.blocks.18.mlp.down_proj", "visual.blocks.18.mlp.gate_proj", "visual.blocks.18.mlp.up_proj", "visual.blocks.19.attn.proj", "visual.blocks.19.attn.qkv_proj", "visual.blocks.19.mlp.down_proj", "visual.blocks.19.mlp.gate_proj", "visual.blocks.19.mlp.up_proj", "visual.blocks.2.attn.proj", "visual.blocks.2.attn.qkv_proj", "visual.blocks.2.mlp.down_proj", "visual.blocks.2.mlp.gate_proj", "visual.blocks.2.mlp.up_proj", "visual.blocks.20.attn.proj", "visual.blocks.20.attn.qkv_proj", "visual.blocks.20.mlp.down_proj", "visual.blocks.20.mlp.gate_proj", "visual.blocks.20.mlp.up_proj", "visual.blocks.21.attn.proj", "visual.blocks.21.attn.qkv_proj", "visual.blocks.21.mlp.down_proj", "visual.blocks.21.mlp.gate_proj", "visual.blocks.21.mlp.up_proj", "visual.blocks.22.attn.proj", "visual.blocks.22.attn.qkv_proj", "visual.blocks.22.mlp.down_proj", "visual.blocks.22.mlp.gate_proj", "visual.blocks.22.mlp.up_proj", "visual.blocks.23.attn.proj", "visual.blocks.23.attn.qkv_proj", "visual.blocks.23.mlp.down_proj", "visual.blocks.23.mlp.gate_proj", "visual.blocks.23.mlp.up_proj", "visual.blocks.3.attn.proj", "visual.blocks.3.attn.qkv_proj", "visual.blocks.3.mlp.down_proj", "visual.blocks.3.mlp.gate_proj", "visual.blocks.3.mlp.up_proj", "visual.blocks.4.attn.proj", "visual.blocks.4.attn.qkv_proj", "visual.blocks.4.mlp.down_proj", "visual.blocks.4.mlp.gate_proj", "visual.blocks.4.mlp.up_proj", "visual.blocks.5.attn.proj", "visual.blocks.5.attn.qkv_proj", "visual.blocks.5.mlp.down_proj", "visual.blocks.5.mlp.gate_proj", "visual.blocks.5.mlp.up_proj", "visual.blocks.6.attn.proj", "visual.blocks.6.attn.qkv_proj", "visual.blocks.6.mlp.down_proj", "visual.blocks.6.mlp.gate_proj", "visual.blocks.6.mlp.up_proj", "visual.blocks.7.attn.proj", "visual.blocks.7.attn.qkv_proj", "visual.blocks.7.mlp.down_proj", "visual.blocks.7.mlp.gate_proj", "visual.blocks.7.mlp.up_proj", "visual.blocks.8.attn.proj", "visual.blocks.8.attn.qkv_proj", "visual.blocks.8.mlp.down_proj", "visual.blocks.8.mlp.gate_proj", "visual.blocks.8.mlp.up_proj", "visual.blocks.9.attn.proj", "visual.blocks.9.attn.qkv_proj", "visual.blocks.9.mlp.down_proj", "visual.blocks.9.mlp.gate_proj", "visual.blocks.9.mlp.up_proj", "visual.downsample", "visual.merger.down_proj", "visual.merger.gate_proj", "visual.merger.proj", "visual.merger.up_proj", "model.layers.45.enorm", "model.layers.45.hnorm", "model.layers.45.input_layernorm", "model.layers.45.mlp.experts.0.down_proj", "model.layers.45.mlp.experts.0.gate_proj", "model.layers.45.mlp.experts.0.up_proj", "model.layers.45.mlp.experts.1.down_proj", "model.layers.45.mlp.experts.1.gate_proj", "model.layers.45.mlp.experts.1.up_proj", "model.layers.45.mlp.experts.10.down_proj", "model.layers.45.mlp.experts.10.gate_proj", "model.layers.45.mlp.experts.10.up_proj", "model.layers.45.mlp.experts.100.down_proj", "model.layers.45.mlp.experts.100.gate_proj", "model.layers.45.mlp.experts.100.up_proj", "model.layers.45.mlp.experts.101.down_proj", "model.layers.45.mlp.experts.101.gate_proj", "model.layers.45.mlp.experts.101.up_proj", "model.layers.45.mlp.experts.102.down_proj", "model.layers.45.mlp.experts.102.gate_proj", "model.layers.45.mlp.experts.102.up_proj", "model.layers.45.mlp.experts.103.down_proj", "model.layers.45.mlp.experts.103.gate_proj", "model.layers.45.mlp.experts.103.up_proj", "model.layers.45.mlp.experts.104.down_proj", "model.layers.45.mlp.experts.104.gate_proj", "model.layers.45.mlp.experts.104.up_proj", "model.layers.45.mlp.experts.105.down_proj", "model.layers.45.mlp.experts.105.gate_proj", "model.layers.45.mlp.experts.105.up_proj", "model.layers.45.mlp.experts.106.down_proj", "model.layers.45.mlp.experts.106.gate_proj", "model.layers.45.mlp.experts.106.up_proj", "model.layers.45.mlp.experts.107.down_proj", "model.layers.45.mlp.experts.107.gate_proj", "model.layers.45.mlp.experts.107.up_proj", "model.layers.45.mlp.experts.108.down_proj", "model.layers.45.mlp.experts.108.gate_proj", "model.layers.45.mlp.experts.108.up_proj", "model.layers.45.mlp.experts.109.down_proj", "model.layers.45.mlp.experts.109.gate_proj", "model.layers.45.mlp.experts.109.up_proj", "model.layers.45.mlp.experts.11.down_proj", "model.layers.45.mlp.experts.11.gate_proj", "model.layers.45.mlp.experts.11.up_proj", "model.layers.45.mlp.experts.110.down_proj", "model.layers.45.mlp.experts.110.gate_proj", "model.layers.45.mlp.experts.110.up_proj", "model.layers.45.mlp.experts.111.down_proj", "model.layers.45.mlp.experts.111.gate_proj", "model.layers.45.mlp.experts.111.up_proj", "model.layers.45.mlp.experts.112.down_proj", "model.layers.45.mlp.experts.112.gate_proj", "model.layers.45.mlp.experts.112.up_proj", "model.layers.45.mlp.experts.113.down_proj", "model.layers.45.mlp.experts.113.gate_proj", "model.layers.45.mlp.experts.113.up_proj", "model.layers.45.mlp.experts.114.down_proj", "model.layers.45.mlp.experts.114.gate_proj", "model.layers.45.mlp.experts.114.up_proj", "model.layers.45.mlp.experts.115.down_proj", "model.layers.45.mlp.experts.115.gate_proj", "model.layers.45.mlp.experts.115.up_proj", "model.layers.45.mlp.experts.116.down_proj", "model.layers.45.mlp.experts.116.gate_proj", "model.layers.45.mlp.experts.116.up_proj", "model.layers.45.mlp.experts.117.down_proj", "model.layers.45.mlp.experts.117.gate_proj", "model.layers.45.mlp.experts.117.up_proj", "model.layers.45.mlp.experts.118.down_proj", "model.layers.45.mlp.experts.118.gate_proj", "model.layers.45.mlp.experts.118.up_proj", "model.layers.45.mlp.experts.119.down_proj", "model.layers.45.mlp.experts.119.gate_proj", "model.layers.45.mlp.experts.119.up_proj", "model.layers.45.mlp.experts.12.down_proj", "model.layers.45.mlp.experts.12.gate_proj", "model.layers.45.mlp.experts.12.up_proj", "model.layers.45.mlp.experts.120.down_proj", "model.layers.45.mlp.experts.120.gate_proj", "model.layers.45.mlp.experts.120.up_proj", "model.layers.45.mlp.experts.121.down_proj", "model.layers.45.mlp.experts.121.gate_proj", "model.layers.45.mlp.experts.121.up_proj", "model.layers.45.mlp.experts.122.down_proj", "model.layers.45.mlp.experts.122.gate_proj", "model.layers.45.mlp.experts.122.up_proj", "model.layers.45.mlp.experts.123.down_proj", "model.layers.45.mlp.experts.123.gate_proj", "model.layers.45.mlp.experts.123.up_proj", "model.layers.45.mlp.experts.124.down_proj", "model.layers.45.mlp.experts.124.gate_proj", "model.layers.45.mlp.experts.124.up_proj", "model.layers.45.mlp.experts.125.down_proj", "model.layers.45.mlp.experts.125.gate_proj", "model.layers.45.mlp.experts.125.up_proj", "model.layers.45.mlp.experts.126.down_proj", "model.layers.45.mlp.experts.126.gate_proj", "model.layers.45.mlp.experts.126.up_proj", "model.layers.45.mlp.experts.127.down_proj", "model.layers.45.mlp.experts.127.gate_proj", "model.layers.45.mlp.experts.127.up_proj", "model.layers.45.mlp.experts.128.down_proj", "model.layers.45.mlp.experts.128.gate_proj", "model.layers.45.mlp.experts.128.up_proj", "model.layers.45.mlp.experts.129.down_proj", "model.layers.45.mlp.experts.129.gate_proj", "model.layers.45.mlp.experts.129.up_proj", "model.layers.45.mlp.experts.13.down_proj", "model.layers.45.mlp.experts.13.gate_proj", "model.layers.45.mlp.experts.13.up_proj", "model.layers.45.mlp.experts.130.down_proj", "model.layers.45.mlp.experts.130.gate_proj", "model.layers.45.mlp.experts.130.up_proj", "model.layers.45.mlp.experts.131.down_proj", "model.layers.45.mlp.experts.131.gate_proj", "model.layers.45.mlp.experts.131.up_proj", "model.layers.45.mlp.experts.132.down_proj", "model.layers.45.mlp.experts.132.gate_proj", "model.layers.45.mlp.experts.132.up_proj", "model.layers.45.mlp.experts.133.down_proj", "model.layers.45.mlp.experts.133.gate_proj", "model.layers.45.mlp.experts.133.up_proj", "model.layers.45.mlp.experts.134.down_proj", "model.layers.45.mlp.experts.134.gate_proj", "model.layers.45.mlp.experts.134.up_proj", "model.layers.45.mlp.experts.135.down_proj", "model.layers.45.mlp.experts.135.gate_proj", "model.layers.45.mlp.experts.135.up_proj", "model.layers.45.mlp.experts.136.down_proj", "model.layers.45.mlp.experts.136.gate_proj", "model.layers.45.mlp.experts.136.up_proj", "model.layers.45.mlp.experts.137.down_proj", "model.layers.45.mlp.experts.137.gate_proj", "model.layers.45.mlp.experts.137.up_proj", "model.layers.45.mlp.experts.138.down_proj", "model.layers.45.mlp.experts.138.gate_proj", "model.layers.45.mlp.experts.138.up_proj", "model.layers.45.mlp.experts.139.down_proj", "model.layers.45.mlp.experts.139.gate_proj", "model.layers.45.mlp.experts.139.up_proj", "model.layers.45.mlp.experts.14.down_proj", "model.layers.45.mlp.experts.14.gate_proj", "model.layers.45.mlp.experts.14.up_proj", "model.layers.45.mlp.experts.140.down_proj", "model.layers.45.mlp.experts.140.gate_proj", "model.layers.45.mlp.experts.140.up_proj", "model.layers.45.mlp.experts.141.down_proj", "model.layers.45.mlp.experts.141.gate_proj", "model.layers.45.mlp.experts.141.up_proj", "model.layers.45.mlp.experts.142.down_proj", "model.layers.45.mlp.experts.142.gate_proj", "model.layers.45.mlp.experts.142.up_proj", "model.layers.45.mlp.experts.143.down_proj", "model.layers.45.mlp.experts.143.gate_proj", "model.layers.45.mlp.experts.143.up_proj", "model.layers.45.mlp.experts.144.down_proj", "model.layers.45.mlp.experts.144.gate_proj", "model.layers.45.mlp.experts.144.up_proj", "model.layers.45.mlp.experts.145.down_proj", "model.layers.45.mlp.experts.145.gate_proj", "model.layers.45.mlp.experts.145.up_proj", "model.layers.45.mlp.experts.146.down_proj", "model.layers.45.mlp.experts.146.gate_proj", "model.layers.45.mlp.experts.146.up_proj", "model.layers.45.mlp.experts.147.down_proj", "model.layers.45.mlp.experts.147.gate_proj", "model.layers.45.mlp.experts.147.up_proj", "model.layers.45.mlp.experts.148.down_proj", "model.layers.45.mlp.experts.148.gate_proj", "model.layers.45.mlp.experts.148.up_proj", "model.layers.45.mlp.experts.149.down_proj", "model.layers.45.mlp.experts.149.gate_proj", "model.layers.45.mlp.experts.149.up_proj", "model.layers.45.mlp.experts.15.down_proj", "model.layers.45.mlp.experts.15.gate_proj", "model.layers.45.mlp.experts.15.up_proj", "model.layers.45.mlp.experts.150.down_proj", "model.layers.45.mlp.experts.150.gate_proj", "model.layers.45.mlp.experts.150.up_proj", "model.layers.45.mlp.experts.151.down_proj", "model.layers.45.mlp.experts.151.gate_proj", "model.layers.45.mlp.experts.151.up_proj", "model.layers.45.mlp.experts.152.down_proj", "model.layers.45.mlp.experts.152.gate_proj", "model.layers.45.mlp.experts.152.up_proj", "model.layers.45.mlp.experts.153.down_proj", "model.layers.45.mlp.experts.153.gate_proj", "model.layers.45.mlp.experts.153.up_proj", "model.layers.45.mlp.experts.154.down_proj", "model.layers.45.mlp.experts.154.gate_proj", "model.layers.45.mlp.experts.154.up_proj", "model.layers.45.mlp.experts.155.down_proj", "model.layers.45.mlp.experts.155.gate_proj", "model.layers.45.mlp.experts.155.up_proj", "model.layers.45.mlp.experts.156.down_proj", "model.layers.45.mlp.experts.156.gate_proj", "model.layers.45.mlp.experts.156.up_proj", "model.layers.45.mlp.experts.157.down_proj", "model.layers.45.mlp.experts.157.gate_proj", "model.layers.45.mlp.experts.157.up_proj", "model.layers.45.mlp.experts.158.down_proj", "model.layers.45.mlp.experts.158.gate_proj", "model.layers.45.mlp.experts.158.up_proj", "model.layers.45.mlp.experts.159.down_proj", "model.layers.45.mlp.experts.159.gate_proj", "model.layers.45.mlp.experts.159.up_proj", "model.layers.45.mlp.experts.16.down_proj", "model.layers.45.mlp.experts.16.gate_proj", "model.layers.45.mlp.experts.16.up_proj", "model.layers.45.mlp.experts.160.down_proj", "model.layers.45.mlp.experts.160.gate_proj", "model.layers.45.mlp.experts.160.up_proj", "model.layers.45.mlp.experts.161.down_proj", "model.layers.45.mlp.experts.161.gate_proj", "model.layers.45.mlp.experts.161.up_proj", "model.layers.45.mlp.experts.162.down_proj", "model.layers.45.mlp.experts.162.gate_proj", "model.layers.45.mlp.experts.162.up_proj", "model.layers.45.mlp.experts.163.down_proj", "model.layers.45.mlp.experts.163.gate_proj", "model.layers.45.mlp.experts.163.up_proj", "model.layers.45.mlp.experts.164.down_proj", "model.layers.45.mlp.experts.164.gate_proj", "model.layers.45.mlp.experts.164.up_proj", "model.layers.45.mlp.experts.165.down_proj", "model.layers.45.mlp.experts.165.gate_proj", "model.layers.45.mlp.experts.165.up_proj", "model.layers.45.mlp.experts.166.down_proj", "model.layers.45.mlp.experts.166.gate_proj", "model.layers.45.mlp.experts.166.up_proj", "model.layers.45.mlp.experts.167.down_proj", "model.layers.45.mlp.experts.167.gate_proj", "model.layers.45.mlp.experts.167.up_proj", "model.layers.45.mlp.experts.168.down_proj", "model.layers.45.mlp.experts.168.gate_proj", "model.layers.45.mlp.experts.168.up_proj", "model.layers.45.mlp.experts.169.down_proj", "model.layers.45.mlp.experts.169.gate_proj", "model.layers.45.mlp.experts.169.up_proj", "model.layers.45.mlp.experts.17.down_proj", "model.layers.45.mlp.experts.17.gate_proj", "model.layers.45.mlp.experts.17.up_proj", "model.layers.45.mlp.experts.170.down_proj", "model.layers.45.mlp.experts.170.gate_proj", "model.layers.45.mlp.experts.170.up_proj", "model.layers.45.mlp.experts.171.down_proj", "model.layers.45.mlp.experts.171.gate_proj", "model.layers.45.mlp.experts.171.up_proj", "model.layers.45.mlp.experts.172.down_proj", "model.layers.45.mlp.experts.172.gate_proj", "model.layers.45.mlp.experts.172.up_proj", "model.layers.45.mlp.experts.173.down_proj", "model.layers.45.mlp.experts.173.gate_proj", "model.layers.45.mlp.experts.173.up_proj", "model.layers.45.mlp.experts.174.down_proj", "model.layers.45.mlp.experts.174.gate_proj", "model.layers.45.mlp.experts.174.up_proj", "model.layers.45.mlp.experts.175.down_proj", "model.layers.45.mlp.experts.175.gate_proj", "model.layers.45.mlp.experts.175.up_proj", "model.layers.45.mlp.experts.176.down_proj", "model.layers.45.mlp.experts.176.gate_proj", "model.layers.45.mlp.experts.176.up_proj", "model.layers.45.mlp.experts.177.down_proj", "model.layers.45.mlp.experts.177.gate_proj", "model.layers.45.mlp.experts.177.up_proj", "model.layers.45.mlp.experts.178.down_proj", "model.layers.45.mlp.experts.178.gate_proj", "model.layers.45.mlp.experts.178.up_proj", "model.layers.45.mlp.experts.179.down_proj", "model.layers.45.mlp.experts.179.gate_proj", "model.layers.45.mlp.experts.179.up_proj", "model.layers.45.mlp.experts.18.down_proj", "model.layers.45.mlp.experts.18.gate_proj", "model.layers.45.mlp.experts.18.up_proj", "model.layers.45.mlp.experts.180.down_proj", "model.layers.45.mlp.experts.180.gate_proj", "model.layers.45.mlp.experts.180.up_proj", "model.layers.45.mlp.experts.181.down_proj", "model.layers.45.mlp.experts.181.gate_proj", "model.layers.45.mlp.experts.181.up_proj", "model.layers.45.mlp.experts.182.down_proj", "model.layers.45.mlp.experts.182.gate_proj", "model.layers.45.mlp.experts.182.up_proj", "model.layers.45.mlp.experts.183.down_proj", "model.layers.45.mlp.experts.183.gate_proj", "model.layers.45.mlp.experts.183.up_proj", "model.layers.45.mlp.experts.184.down_proj", "model.layers.45.mlp.experts.184.gate_proj", "model.layers.45.mlp.experts.184.up_proj", "model.layers.45.mlp.experts.185.down_proj", "model.layers.45.mlp.experts.185.gate_proj", "model.layers.45.mlp.experts.185.up_proj", "model.layers.45.mlp.experts.186.down_proj", "model.layers.45.mlp.experts.186.gate_proj", "model.layers.45.mlp.experts.186.up_proj", "model.layers.45.mlp.experts.187.down_proj", "model.layers.45.mlp.experts.187.gate_proj", "model.layers.45.mlp.experts.187.up_proj", "model.layers.45.mlp.experts.188.down_proj", "model.layers.45.mlp.experts.188.gate_proj", "model.layers.45.mlp.experts.188.up_proj", "model.layers.45.mlp.experts.189.down_proj", "model.layers.45.mlp.experts.189.gate_proj", "model.layers.45.mlp.experts.189.up_proj", "model.layers.45.mlp.experts.19.down_proj", "model.layers.45.mlp.experts.19.gate_proj", "model.layers.45.mlp.experts.19.up_proj", "model.layers.45.mlp.experts.190.down_proj", "model.layers.45.mlp.experts.190.gate_proj", "model.layers.45.mlp.experts.190.up_proj", "model.layers.45.mlp.experts.191.down_proj", "model.layers.45.mlp.experts.191.gate_proj", "model.layers.45.mlp.experts.191.up_proj", "model.layers.45.mlp.experts.192.down_proj", "model.layers.45.mlp.experts.192.gate_proj", "model.layers.45.mlp.experts.192.up_proj", "model.layers.45.mlp.experts.193.down_proj", "model.layers.45.mlp.experts.193.gate_proj", "model.layers.45.mlp.experts.193.up_proj", "model.layers.45.mlp.experts.194.down_proj", "model.layers.45.mlp.experts.194.gate_proj", "model.layers.45.mlp.experts.194.up_proj", "model.layers.45.mlp.experts.195.down_proj", "model.layers.45.mlp.experts.195.gate_proj", "model.layers.45.mlp.experts.195.up_proj", "model.layers.45.mlp.experts.196.down_proj", "model.layers.45.mlp.experts.196.gate_proj", "model.layers.45.mlp.experts.196.up_proj", "model.layers.45.mlp.experts.197.down_proj", "model.layers.45.mlp.experts.197.gate_proj", "model.layers.45.mlp.experts.197.up_proj", "model.layers.45.mlp.experts.198.down_proj", "model.layers.45.mlp.experts.198.gate_proj", "model.layers.45.mlp.experts.198.up_proj", "model.layers.45.mlp.experts.199.down_proj", "model.layers.45.mlp.experts.199.gate_proj", "model.layers.45.mlp.experts.199.up_proj", "model.layers.45.mlp.experts.2.down_proj", "model.layers.45.mlp.experts.2.gate_proj", "model.layers.45.mlp.experts.2.up_proj", "model.layers.45.mlp.experts.20.down_proj", "model.layers.45.mlp.experts.20.gate_proj", "model.layers.45.mlp.experts.20.up_proj", "model.layers.45.mlp.experts.200.down_proj", "model.layers.45.mlp.experts.200.gate_proj", "model.layers.45.mlp.experts.200.up_proj", "model.layers.45.mlp.experts.201.down_proj", "model.layers.45.mlp.experts.201.gate_proj", "model.layers.45.mlp.experts.201.up_proj", "model.layers.45.mlp.experts.202.down_proj", "model.layers.45.mlp.experts.202.gate_proj", "model.layers.45.mlp.experts.202.up_proj", "model.layers.45.mlp.experts.203.down_proj", "model.layers.45.mlp.experts.203.gate_proj", "model.layers.45.mlp.experts.203.up_proj", "model.layers.45.mlp.experts.204.down_proj", "model.layers.45.mlp.experts.204.gate_proj", "model.layers.45.mlp.experts.204.up_proj", "model.layers.45.mlp.experts.205.down_proj", "model.layers.45.mlp.experts.205.gate_proj", "model.layers.45.mlp.experts.205.up_proj", "model.layers.45.mlp.experts.206.down_proj", "model.layers.45.mlp.experts.206.gate_proj", "model.layers.45.mlp.experts.206.up_proj", "model.layers.45.mlp.experts.207.down_proj", "model.layers.45.mlp.experts.207.gate_proj", "model.layers.45.mlp.experts.207.up_proj", "model.layers.45.mlp.experts.208.down_proj", "model.layers.45.mlp.experts.208.gate_proj", "model.layers.45.mlp.experts.208.up_proj", "model.layers.45.mlp.experts.209.down_proj", "model.layers.45.mlp.experts.209.gate_proj", "model.layers.45.mlp.experts.209.up_proj", "model.layers.45.mlp.experts.21.down_proj", "model.layers.45.mlp.experts.21.gate_proj", "model.layers.45.mlp.experts.21.up_proj", "model.layers.45.mlp.experts.210.down_proj", "model.layers.45.mlp.experts.210.gate_proj", "model.layers.45.mlp.experts.210.up_proj", "model.layers.45.mlp.experts.211.down_proj", "model.layers.45.mlp.experts.211.gate_proj", "model.layers.45.mlp.experts.211.up_proj", "model.layers.45.mlp.experts.212.down_proj", "model.layers.45.mlp.experts.212.gate_proj", "model.layers.45.mlp.experts.212.up_proj", "model.layers.45.mlp.experts.213.down_proj", "model.layers.45.mlp.experts.213.gate_proj", "model.layers.45.mlp.experts.213.up_proj", "model.layers.45.mlp.experts.214.down_proj", "model.layers.45.mlp.experts.214.gate_proj", "model.layers.45.mlp.experts.214.up_proj", "model.layers.45.mlp.experts.215.down_proj", "model.layers.45.mlp.experts.215.gate_proj", "model.layers.45.mlp.experts.215.up_proj", "model.layers.45.mlp.experts.216.down_proj", "model.layers.45.mlp.experts.216.gate_proj", "model.layers.45.mlp.experts.216.up_proj", "model.layers.45.mlp.experts.217.down_proj", "model.layers.45.mlp.experts.217.gate_proj", "model.layers.45.mlp.experts.217.up_proj", "model.layers.45.mlp.experts.218.down_proj", "model.layers.45.mlp.experts.218.gate_proj", "model.layers.45.mlp.experts.218.up_proj", "model.layers.45.mlp.experts.219.down_proj", "model.layers.45.mlp.experts.219.gate_proj", "model.layers.45.mlp.experts.219.up_proj", "model.layers.45.mlp.experts.22.down_proj", "model.layers.45.mlp.experts.22.gate_proj", "model.layers.45.mlp.experts.22.up_proj", "model.layers.45.mlp.experts.220.down_proj", "model.layers.45.mlp.experts.220.gate_proj", "model.layers.45.mlp.experts.220.up_proj", "model.layers.45.mlp.experts.221.down_proj", "model.layers.45.mlp.experts.221.gate_proj", "model.layers.45.mlp.experts.221.up_proj", "model.layers.45.mlp.experts.222.down_proj", "model.layers.45.mlp.experts.222.gate_proj", "model.layers.45.mlp.experts.222.up_proj", "model.layers.45.mlp.experts.223.down_proj", "model.layers.45.mlp.experts.223.gate_proj", "model.layers.45.mlp.experts.223.up_proj", "model.layers.45.mlp.experts.224.down_proj", "model.layers.45.mlp.experts.224.gate_proj", "model.layers.45.mlp.experts.224.up_proj", "model.layers.45.mlp.experts.225.down_proj", "model.layers.45.mlp.experts.225.gate_proj", "model.layers.45.mlp.experts.225.up_proj", "model.layers.45.mlp.experts.226.down_proj", "model.layers.45.mlp.experts.226.gate_proj", "model.layers.45.mlp.experts.226.up_proj", "model.layers.45.mlp.experts.227.down_proj", "model.layers.45.mlp.experts.227.gate_proj", "model.layers.45.mlp.experts.227.up_proj", "model.layers.45.mlp.experts.228.down_proj", "model.layers.45.mlp.experts.228.gate_proj", "model.layers.45.mlp.experts.228.up_proj", "model.layers.45.mlp.experts.229.down_proj", "model.layers.45.mlp.experts.229.gate_proj", "model.layers.45.mlp.experts.229.up_proj", "model.layers.45.mlp.experts.23.down_proj", "model.layers.45.mlp.experts.23.gate_proj", "model.layers.45.mlp.experts.23.up_proj", "model.layers.45.mlp.experts.230.down_proj", "model.layers.45.mlp.experts.230.gate_proj", "model.layers.45.mlp.experts.230.up_proj", "model.layers.45.mlp.experts.231.down_proj", "model.layers.45.mlp.experts.231.gate_proj", "model.layers.45.mlp.experts.231.up_proj", "model.layers.45.mlp.experts.232.down_proj", "model.layers.45.mlp.experts.232.gate_proj", "model.layers.45.mlp.experts.232.up_proj", "model.layers.45.mlp.experts.233.down_proj", "model.layers.45.mlp.experts.233.gate_proj", "model.layers.45.mlp.experts.233.up_proj", "model.layers.45.mlp.experts.234.down_proj", "model.layers.45.mlp.experts.234.gate_proj", "model.layers.45.mlp.experts.234.up_proj", "model.layers.45.mlp.experts.235.down_proj", "model.layers.45.mlp.experts.235.gate_proj", "model.layers.45.mlp.experts.235.up_proj", "model.layers.45.mlp.experts.236.down_proj", "model.layers.45.mlp.experts.236.gate_proj", "model.layers.45.mlp.experts.236.up_proj", "model.layers.45.mlp.experts.237.down_proj", "model.layers.45.mlp.experts.237.gate_proj", "model.layers.45.mlp.experts.237.up_proj", "model.layers.45.mlp.experts.238.down_proj", "model.layers.45.mlp.experts.238.gate_proj", "model.layers.45.mlp.experts.238.up_proj", "model.layers.45.mlp.experts.239.down_proj", "model.layers.45.mlp.experts.239.gate_proj", "model.layers.45.mlp.experts.239.up_proj", "model.layers.45.mlp.experts.24.down_proj", "model.layers.45.mlp.experts.24.gate_proj", "model.layers.45.mlp.experts.24.up_proj", "model.layers.45.mlp.experts.240.down_proj", "model.layers.45.mlp.experts.240.gate_proj", "model.layers.45.mlp.experts.240.up_proj", "model.layers.45.mlp.experts.241.down_proj", "model.layers.45.mlp.experts.241.gate_proj", "model.layers.45.mlp.experts.241.up_proj", "model.layers.45.mlp.experts.242.down_proj", "model.layers.45.mlp.experts.242.gate_proj", "model.layers.45.mlp.experts.242.up_proj", "model.layers.45.mlp.experts.243.down_proj", "model.layers.45.mlp.experts.243.gate_proj", "model.layers.45.mlp.experts.243.up_proj", "model.layers.45.mlp.experts.244.down_proj", "model.layers.45.mlp.experts.244.gate_proj", "model.layers.45.mlp.experts.244.up_proj", "model.layers.45.mlp.experts.245.down_proj", "model.layers.45.mlp.experts.245.gate_proj", "model.layers.45.mlp.experts.245.up_proj", "model.layers.45.mlp.experts.246.down_proj", "model.layers.45.mlp.experts.246.gate_proj", "model.layers.45.mlp.experts.246.up_proj", "model.layers.45.mlp.experts.247.down_proj", "model.layers.45.mlp.experts.247.gate_proj", "model.layers.45.mlp.experts.247.up_proj", "model.layers.45.mlp.experts.248.down_proj", "model.layers.45.mlp.experts.248.gate_proj", "model.layers.45.mlp.experts.248.up_proj", "model.layers.45.mlp.experts.249.down_proj", "model.layers.45.mlp.experts.249.gate_proj", "model.layers.45.mlp.experts.249.up_proj", "model.layers.45.mlp.experts.25.down_proj", "model.layers.45.mlp.experts.25.gate_proj", "model.layers.45.mlp.experts.25.up_proj", "model.layers.45.mlp.experts.250.down_proj", "model.layers.45.mlp.experts.250.gate_proj", "model.layers.45.mlp.experts.250.up_proj", "model.layers.45.mlp.experts.251.down_proj", "model.layers.45.mlp.experts.251.gate_proj", "model.layers.45.mlp.experts.251.up_proj", "model.layers.45.mlp.experts.252.down_proj", "model.layers.45.mlp.experts.252.gate_proj", "model.layers.45.mlp.experts.252.up_proj", "model.layers.45.mlp.experts.253.down_proj", "model.layers.45.mlp.experts.253.gate_proj", "model.layers.45.mlp.experts.253.up_proj", "model.layers.45.mlp.experts.254.down_proj", "model.layers.45.mlp.experts.254.gate_proj", "model.layers.45.mlp.experts.254.up_proj", "model.layers.45.mlp.experts.255.down_proj", "model.layers.45.mlp.experts.255.gate_proj", "model.layers.45.mlp.experts.255.up_proj", "model.layers.45.mlp.experts.256.down_proj", "model.layers.45.mlp.experts.256.gate_proj", "model.layers.45.mlp.experts.256.up_proj", "model.layers.45.mlp.experts.257.down_proj", "model.layers.45.mlp.experts.257.gate_proj", "model.layers.45.mlp.experts.257.up_proj", "model.layers.45.mlp.experts.258.down_proj", "model.layers.45.mlp.experts.258.gate_proj", "model.layers.45.mlp.experts.258.up_proj", "model.layers.45.mlp.experts.259.down_proj", "model.layers.45.mlp.experts.259.gate_proj", "model.layers.45.mlp.experts.259.up_proj", "model.layers.45.mlp.experts.26.down_proj", "model.layers.45.mlp.experts.26.gate_proj", "model.layers.45.mlp.experts.26.up_proj", "model.layers.45.mlp.experts.260.down_proj", "model.layers.45.mlp.experts.260.gate_proj", "model.layers.45.mlp.experts.260.up_proj", "model.layers.45.mlp.experts.261.down_proj", "model.layers.45.mlp.experts.261.gate_proj", "model.layers.45.mlp.experts.261.up_proj", "model.layers.45.mlp.experts.262.down_proj", "model.layers.45.mlp.experts.262.gate_proj", "model.layers.45.mlp.experts.262.up_proj", "model.layers.45.mlp.experts.263.down_proj", "model.layers.45.mlp.experts.263.gate_proj", "model.layers.45.mlp.experts.263.up_proj", "model.layers.45.mlp.experts.264.down_proj", "model.layers.45.mlp.experts.264.gate_proj", "model.layers.45.mlp.experts.264.up_proj", "model.layers.45.mlp.experts.265.down_proj", "model.layers.45.mlp.experts.265.gate_proj", "model.layers.45.mlp.experts.265.up_proj", "model.layers.45.mlp.experts.266.down_proj", "model.layers.45.mlp.experts.266.gate_proj", "model.layers.45.mlp.experts.266.up_proj", "model.layers.45.mlp.experts.267.down_proj", "model.layers.45.mlp.experts.267.gate_proj", "model.layers.45.mlp.experts.267.up_proj", "model.layers.45.mlp.experts.268.down_proj", "model.layers.45.mlp.experts.268.gate_proj", "model.layers.45.mlp.experts.268.up_proj", "model.layers.45.mlp.experts.269.down_proj", "model.layers.45.mlp.experts.269.gate_proj", "model.layers.45.mlp.experts.269.up_proj", "model.layers.45.mlp.experts.27.down_proj", "model.layers.45.mlp.experts.27.gate_proj", "model.layers.45.mlp.experts.27.up_proj", "model.layers.45.mlp.experts.270.down_proj", "model.layers.45.mlp.experts.270.gate_proj", "model.layers.45.mlp.experts.270.up_proj", "model.layers.45.mlp.experts.271.down_proj", "model.layers.45.mlp.experts.271.gate_proj", "model.layers.45.mlp.experts.271.up_proj", "model.layers.45.mlp.experts.272.down_proj", "model.layers.45.mlp.experts.272.gate_proj", "model.layers.45.mlp.experts.272.up_proj", "model.layers.45.mlp.experts.273.down_proj", "model.layers.45.mlp.experts.273.gate_proj", "model.layers.45.mlp.experts.273.up_proj", "model.layers.45.mlp.experts.274.down_proj", "model.layers.45.mlp.experts.274.gate_proj", "model.layers.45.mlp.experts.274.up_proj", "model.layers.45.mlp.experts.275.down_proj", "model.layers.45.mlp.experts.275.gate_proj", "model.layers.45.mlp.experts.275.up_proj", "model.layers.45.mlp.experts.276.down_proj", "model.layers.45.mlp.experts.276.gate_proj", "model.layers.45.mlp.experts.276.up_proj", "model.layers.45.mlp.experts.277.down_proj", "model.layers.45.mlp.experts.277.gate_proj", "model.layers.45.mlp.experts.277.up_proj", "model.layers.45.mlp.experts.278.down_proj", "model.layers.45.mlp.experts.278.gate_proj", "model.layers.45.mlp.experts.278.up_proj", "model.layers.45.mlp.experts.279.down_proj", "model.layers.45.mlp.experts.279.gate_proj", "model.layers.45.mlp.experts.279.up_proj", "model.layers.45.mlp.experts.28.down_proj", "model.layers.45.mlp.experts.28.gate_proj", "model.layers.45.mlp.experts.28.up_proj", "model.layers.45.mlp.experts.280.down_proj", "model.layers.45.mlp.experts.280.gate_proj", "model.layers.45.mlp.experts.280.up_proj", "model.layers.45.mlp.experts.281.down_proj", "model.layers.45.mlp.experts.281.gate_proj", "model.layers.45.mlp.experts.281.up_proj", "model.layers.45.mlp.experts.282.down_proj", "model.layers.45.mlp.experts.282.gate_proj", "model.layers.45.mlp.experts.282.up_proj", "model.layers.45.mlp.experts.283.down_proj", "model.layers.45.mlp.experts.283.gate_proj", "model.layers.45.mlp.experts.283.up_proj", "model.layers.45.mlp.experts.284.down_proj", "model.layers.45.mlp.experts.284.gate_proj", "model.layers.45.mlp.experts.284.up_proj", "model.layers.45.mlp.experts.285.down_proj", "model.layers.45.mlp.experts.285.gate_proj", "model.layers.45.mlp.experts.285.up_proj", "model.layers.45.mlp.experts.286.down_proj", "model.layers.45.mlp.experts.286.gate_proj", "model.layers.45.mlp.experts.286.up_proj", "model.layers.45.mlp.experts.287.down_proj", "model.layers.45.mlp.experts.287.gate_proj", "model.layers.45.mlp.experts.287.up_proj", "model.layers.45.mlp.experts.29.down_proj", "model.layers.45.mlp.experts.29.gate_proj", "model.layers.45.mlp.experts.29.up_proj", "model.layers.45.mlp.experts.3.down_proj", "model.layers.45.mlp.experts.3.gate_proj", "model.layers.45.mlp.experts.3.up_proj", "model.layers.45.mlp.experts.30.down_proj", "model.layers.45.mlp.experts.30.gate_proj", "model.layers.45.mlp.experts.30.up_proj", "model.layers.45.mlp.experts.31.down_proj", "model.layers.45.mlp.experts.31.gate_proj", "model.layers.45.mlp.experts.31.up_proj", "model.layers.45.mlp.experts.32.down_proj", "model.layers.45.mlp.experts.32.gate_proj", "model.layers.45.mlp.experts.32.up_proj", "model.layers.45.mlp.experts.33.down_proj", "model.layers.45.mlp.experts.33.gate_proj", "model.layers.45.mlp.experts.33.up_proj", "model.layers.45.mlp.experts.34.down_proj", "model.layers.45.mlp.experts.34.gate_proj", "model.layers.45.mlp.experts.34.up_proj", "model.layers.45.mlp.experts.35.down_proj", "model.layers.45.mlp.experts.35.gate_proj", "model.layers.45.mlp.experts.35.up_proj", "model.layers.45.mlp.experts.36.down_proj", "model.layers.45.mlp.experts.36.gate_proj", "model.layers.45.mlp.experts.36.up_proj", "model.layers.45.mlp.experts.37.down_proj", "model.layers.45.mlp.experts.37.gate_proj", "model.layers.45.mlp.experts.37.up_proj", "model.layers.45.mlp.experts.38.down_proj", "model.layers.45.mlp.experts.38.gate_proj", "model.layers.45.mlp.experts.38.up_proj", "model.layers.45.mlp.experts.39.down_proj", "model.layers.45.mlp.experts.39.gate_proj", "model.layers.45.mlp.experts.39.up_proj", "model.layers.45.mlp.experts.4.down_proj", "model.layers.45.mlp.experts.4.gate_proj", "model.layers.45.mlp.experts.4.up_proj", "model.layers.45.mlp.experts.40.down_proj", "model.layers.45.mlp.experts.40.gate_proj", "model.layers.45.mlp.experts.40.up_proj", "model.layers.45.mlp.experts.41.down_proj", "model.layers.45.mlp.experts.41.gate_proj", "model.layers.45.mlp.experts.41.up_proj", "model.layers.45.mlp.experts.42.down_proj", "model.layers.45.mlp.experts.42.gate_proj", "model.layers.45.mlp.experts.42.up_proj", "model.layers.45.mlp.experts.43.down_proj", "model.layers.45.mlp.experts.43.gate_proj", "model.layers.45.mlp.experts.43.up_proj", "model.layers.45.mlp.experts.44.down_proj", "model.layers.45.mlp.experts.44.gate_proj", "model.layers.45.mlp.experts.44.up_proj", "model.layers.45.mlp.experts.45.down_proj", "model.layers.45.mlp.experts.45.gate_proj", "model.layers.45.mlp.experts.45.up_proj", "model.layers.45.mlp.experts.46.down_proj", "model.layers.45.mlp.experts.46.gate_proj", "model.layers.45.mlp.experts.46.up_proj", "model.layers.45.mlp.experts.47.down_proj", "model.layers.45.mlp.experts.47.gate_proj", "model.layers.45.mlp.experts.47.up_proj", "model.layers.45.mlp.experts.48.down_proj", "model.layers.45.mlp.experts.48.gate_proj", "model.layers.45.mlp.experts.48.up_proj", "model.layers.45.mlp.experts.49.down_proj", "model.layers.45.mlp.experts.49.gate_proj", "model.layers.45.mlp.experts.49.up_proj", "model.layers.45.mlp.experts.5.down_proj", "model.layers.45.mlp.experts.5.gate_proj", "model.layers.45.mlp.experts.5.up_proj", "model.layers.45.mlp.experts.50.down_proj", "model.layers.45.mlp.experts.50.gate_proj", "model.layers.45.mlp.experts.50.up_proj", "model.layers.45.mlp.experts.51.down_proj", "model.layers.45.mlp.experts.51.gate_proj", "model.layers.45.mlp.experts.51.up_proj", "model.layers.45.mlp.experts.52.down_proj", "model.layers.45.mlp.experts.52.gate_proj", "model.layers.45.mlp.experts.52.up_proj", "model.layers.45.mlp.experts.53.down_proj", "model.layers.45.mlp.experts.53.gate_proj", "model.layers.45.mlp.experts.53.up_proj", "model.layers.45.mlp.experts.54.down_proj", "model.layers.45.mlp.experts.54.gate_proj", "model.layers.45.mlp.experts.54.up_proj", "model.layers.45.mlp.experts.55.down_proj", "model.layers.45.mlp.experts.55.gate_proj", "model.layers.45.mlp.experts.55.up_proj", "model.layers.45.mlp.experts.56.down_proj", "model.layers.45.mlp.experts.56.gate_proj", "model.layers.45.mlp.experts.56.up_proj", "model.layers.45.mlp.experts.57.down_proj", "model.layers.45.mlp.experts.57.gate_proj", "model.layers.45.mlp.experts.57.up_proj", "model.layers.45.mlp.experts.58.down_proj", "model.layers.45.mlp.experts.58.gate_proj", "model.layers.45.mlp.experts.58.up_proj", "model.layers.45.mlp.experts.59.down_proj", "model.layers.45.mlp.experts.59.gate_proj", "model.layers.45.mlp.experts.59.up_proj", "model.layers.45.mlp.experts.6.down_proj", "model.layers.45.mlp.experts.6.gate_proj", "model.layers.45.mlp.experts.6.up_proj", "model.layers.45.mlp.experts.60.down_proj", "model.layers.45.mlp.experts.60.gate_proj", "model.layers.45.mlp.experts.60.up_proj", "model.layers.45.mlp.experts.61.down_proj", "model.layers.45.mlp.experts.61.gate_proj", "model.layers.45.mlp.experts.61.up_proj", "model.layers.45.mlp.experts.62.down_proj", "model.layers.45.mlp.experts.62.gate_proj", "model.layers.45.mlp.experts.62.up_proj", "model.layers.45.mlp.experts.63.down_proj", "model.layers.45.mlp.experts.63.gate_proj", "model.layers.45.mlp.experts.63.up_proj", "model.layers.45.mlp.experts.64.down_proj", "model.layers.45.mlp.experts.64.gate_proj", "model.layers.45.mlp.experts.64.up_proj", "model.layers.45.mlp.experts.65.down_proj", "model.layers.45.mlp.experts.65.gate_proj", "model.layers.45.mlp.experts.65.up_proj", "model.layers.45.mlp.experts.66.down_proj", "model.layers.45.mlp.experts.66.gate_proj", "model.layers.45.mlp.experts.66.up_proj", "model.layers.45.mlp.experts.67.down_proj", "model.layers.45.mlp.experts.67.gate_proj", "model.layers.45.mlp.experts.67.up_proj", "model.layers.45.mlp.experts.68.down_proj", "model.layers.45.mlp.experts.68.gate_proj", "model.layers.45.mlp.experts.68.up_proj", "model.layers.45.mlp.experts.69.down_proj", "model.layers.45.mlp.experts.69.gate_proj", "model.layers.45.mlp.experts.69.up_proj", "model.layers.45.mlp.experts.7.down_proj", "model.layers.45.mlp.experts.7.gate_proj", "model.layers.45.mlp.experts.7.up_proj", "model.layers.45.mlp.experts.70.down_proj", "model.layers.45.mlp.experts.70.gate_proj", "model.layers.45.mlp.experts.70.up_proj", "model.layers.45.mlp.experts.71.down_proj", "model.layers.45.mlp.experts.71.gate_proj", "model.layers.45.mlp.experts.71.up_proj", "model.layers.45.mlp.experts.72.down_proj", "model.layers.45.mlp.experts.72.gate_proj", "model.layers.45.mlp.experts.72.up_proj", "model.layers.45.mlp.experts.73.down_proj", "model.layers.45.mlp.experts.73.gate_proj", "model.layers.45.mlp.experts.73.up_proj", "model.layers.45.mlp.experts.74.down_proj", "model.layers.45.mlp.experts.74.gate_proj", "model.layers.45.mlp.experts.74.up_proj", "model.layers.45.mlp.experts.75.down_proj", "model.layers.45.mlp.experts.75.gate_proj", "model.layers.45.mlp.experts.75.up_proj", "model.layers.45.mlp.experts.76.down_proj", "model.layers.45.mlp.experts.76.gate_proj", "model.layers.45.mlp.experts.76.up_proj", "model.layers.45.mlp.experts.77.down_proj", "model.layers.45.mlp.experts.77.gate_proj", "model.layers.45.mlp.experts.77.up_proj", "model.layers.45.mlp.experts.78.down_proj", "model.layers.45.mlp.experts.78.gate_proj", "model.layers.45.mlp.experts.78.up_proj", "model.layers.45.mlp.experts.79.down_proj", "model.layers.45.mlp.experts.79.gate_proj", "model.layers.45.mlp.experts.79.up_proj", "model.layers.45.mlp.experts.8.down_proj", "model.layers.45.mlp.experts.8.gate_proj", "model.layers.45.mlp.experts.8.up_proj", "model.layers.45.mlp.experts.80.down_proj", "model.layers.45.mlp.experts.80.gate_proj", "model.layers.45.mlp.experts.80.up_proj", "model.layers.45.mlp.experts.81.down_proj", "model.layers.45.mlp.experts.81.gate_proj", "model.layers.45.mlp.experts.81.up_proj", "model.layers.45.mlp.experts.82.down_proj", "model.layers.45.mlp.experts.82.gate_proj", "model.layers.45.mlp.experts.82.up_proj", "model.layers.45.mlp.experts.83.down_proj", "model.layers.45.mlp.experts.83.gate_proj", "model.layers.45.mlp.experts.83.up_proj", "model.layers.45.mlp.experts.84.down_proj", "model.layers.45.mlp.experts.84.gate_proj", "model.layers.45.mlp.experts.84.up_proj", "model.layers.45.mlp.experts.85.down_proj", "model.layers.45.mlp.experts.85.gate_proj", "model.layers.45.mlp.experts.85.up_proj", "model.layers.45.mlp.experts.86.down_proj", "model.layers.45.mlp.experts.86.gate_proj", "model.layers.45.mlp.experts.86.up_proj", "model.layers.45.mlp.experts.87.down_proj", "model.layers.45.mlp.experts.87.gate_proj", "model.layers.45.mlp.experts.87.up_proj", "model.layers.45.mlp.experts.88.down_proj", "model.layers.45.mlp.experts.88.gate_proj", "model.layers.45.mlp.experts.88.up_proj", "model.layers.45.mlp.experts.89.down_proj", "model.layers.45.mlp.experts.89.gate_proj", "model.layers.45.mlp.experts.89.up_proj", "model.layers.45.mlp.experts.9.down_proj", "model.layers.45.mlp.experts.9.gate_proj", "model.layers.45.mlp.experts.9.up_proj", "model.layers.45.mlp.experts.90.down_proj", "model.layers.45.mlp.experts.90.gate_proj", "model.layers.45.mlp.experts.90.up_proj", "model.layers.45.mlp.experts.91.down_proj", "model.layers.45.mlp.experts.91.gate_proj", "model.layers.45.mlp.experts.91.up_proj", "model.layers.45.mlp.experts.92.down_proj", "model.layers.45.mlp.experts.92.gate_proj", "model.layers.45.mlp.experts.92.up_proj", "model.layers.45.mlp.experts.93.down_proj", "model.layers.45.mlp.experts.93.gate_proj", "model.layers.45.mlp.experts.93.up_proj", "model.layers.45.mlp.experts.94.down_proj", "model.layers.45.mlp.experts.94.gate_proj", "model.layers.45.mlp.experts.94.up_proj", "model.layers.45.mlp.experts.95.down_proj", "model.layers.45.mlp.experts.95.gate_proj", "model.layers.45.mlp.experts.95.up_proj", "model.layers.45.mlp.experts.96.down_proj", "model.layers.45.mlp.experts.96.gate_proj", "model.layers.45.mlp.experts.96.up_proj", "model.layers.45.mlp.experts.97.down_proj", "model.layers.45.mlp.experts.97.gate_proj", "model.layers.45.mlp.experts.97.up_proj", "model.layers.45.mlp.experts.98.down_proj", "model.layers.45.mlp.experts.98.gate_proj", "model.layers.45.mlp.experts.98.up_proj", "model.layers.45.mlp.experts.99.down_proj", "model.layers.45.mlp.experts.99.gate_proj", "model.layers.45.mlp.experts.99.up_proj", "model.layers.45.post_attention_layernorm", "model.layers.45.self_attn.indexer", "model.layers.45.self_attn.indexer.k_norm", "model.layers.45.self_attn.kv_a_layernorm", "model.layers.45.self_attn.q_a_layernorm", "model.layers.45.shared_head.norm", "model.layers.3.mlp.experts", "model.layers.5.mlp.experts", "model.layers.6.mlp.experts", "model.layers.6.mlp.experts.0.down_proj", "model.layers.6.mlp.experts.0.gate_proj", "model.layers.6.mlp.experts.0.up_proj", "model.layers.6.mlp.experts.1.down_proj", "model.layers.6.mlp.experts.1.gate_proj", "model.layers.6.mlp.experts.1.up_proj", "model.layers.6.mlp.experts.10.down_proj", "model.layers.6.mlp.experts.10.gate_proj", "model.layers.6.mlp.experts.10.up_proj", "model.layers.6.mlp.experts.100.down_proj", "model.layers.6.mlp.experts.100.gate_proj", "model.layers.6.mlp.experts.100.up_proj", "model.layers.6.mlp.experts.101.down_proj", "model.layers.6.mlp.experts.101.gate_proj", "model.layers.6.mlp.experts.101.up_proj", "model.layers.6.mlp.experts.102.down_proj", "model.layers.6.mlp.experts.102.gate_proj", "model.layers.6.mlp.experts.102.up_proj", "model.layers.6.mlp.experts.103.down_proj", "model.layers.6.mlp.experts.103.gate_proj", "model.layers.6.mlp.experts.103.up_proj", "model.layers.6.mlp.experts.104.down_proj", "model.layers.6.mlp.experts.104.gate_proj", "model.layers.6.mlp.experts.104.up_proj", "model.layers.6.mlp.experts.105.down_proj", "model.layers.6.mlp.experts.105.gate_proj", "model.layers.6.mlp.experts.105.up_proj", "model.layers.6.mlp.experts.106.down_proj", "model.layers.6.mlp.experts.106.gate_proj", "model.layers.6.mlp.experts.106.up_proj", "model.layers.6.mlp.experts.107.down_proj", "model.layers.6.mlp.experts.107.gate_proj", "model.layers.6.mlp.experts.107.up_proj", "model.layers.6.mlp.experts.108.down_proj", "model.layers.6.mlp.experts.108.gate_proj", "model.layers.6.mlp.experts.108.up_proj", "model.layers.6.mlp.experts.109.down_proj", "model.layers.6.mlp.experts.109.gate_proj", "model.layers.6.mlp.experts.109.up_proj", "model.layers.6.mlp.experts.11.down_proj", "model.layers.6.mlp.experts.11.gate_proj", "model.layers.6.mlp.experts.11.up_proj", "model.layers.6.mlp.experts.110.down_proj", "model.layers.6.mlp.experts.110.gate_proj", "model.layers.6.mlp.experts.110.up_proj", "model.layers.6.mlp.experts.111.down_proj", "model.layers.6.mlp.experts.111.gate_proj", "model.layers.6.mlp.experts.111.up_proj", "model.layers.6.mlp.experts.112.down_proj", "model.layers.6.mlp.experts.112.gate_proj", "model.layers.6.mlp.experts.112.up_proj", "model.layers.6.mlp.experts.113.down_proj", "model.layers.6.mlp.experts.113.gate_proj", "model.layers.6.mlp.experts.113.up_proj", "model.layers.6.mlp.experts.114.down_proj", "model.layers.6.mlp.experts.114.gate_proj", "model.layers.6.mlp.experts.114.up_proj", "model.layers.6.mlp.experts.115.down_proj", "model.layers.6.mlp.experts.115.gate_proj", "model.layers.6.mlp.experts.115.up_proj", "model.layers.6.mlp.experts.116.down_proj", "model.layers.6.mlp.experts.116.gate_proj", "model.layers.6.mlp.experts.116.up_proj", "model.layers.6.mlp.experts.117.down_proj", "model.layers.6.mlp.experts.117.gate_proj", "model.layers.6.mlp.experts.117.up_proj", "model.layers.6.mlp.experts.118.down_proj", "model.layers.6.mlp.experts.118.gate_proj", "model.layers.6.mlp.experts.118.up_proj", "model.layers.6.mlp.experts.119.down_proj", "model.layers.6.mlp.experts.119.gate_proj", "model.layers.6.mlp.experts.119.up_proj", "model.layers.6.mlp.experts.12.down_proj", "model.layers.6.mlp.experts.12.gate_proj", "model.layers.6.mlp.experts.12.up_proj", "model.layers.6.mlp.experts.120.down_proj", "model.layers.6.mlp.experts.120.gate_proj", "model.layers.6.mlp.experts.120.up_proj", "model.layers.6.mlp.experts.121.down_proj", "model.layers.6.mlp.experts.121.gate_proj", "model.layers.6.mlp.experts.121.up_proj", "model.layers.6.mlp.experts.122.down_proj", "model.layers.6.mlp.experts.122.gate_proj", "model.layers.6.mlp.experts.122.up_proj", "model.layers.6.mlp.experts.123.down_proj", "model.layers.6.mlp.experts.123.gate_proj", "model.layers.6.mlp.experts.123.up_proj", "model.layers.6.mlp.experts.124.down_proj", "model.layers.6.mlp.experts.124.gate_proj", "model.layers.6.mlp.experts.124.up_proj", "model.layers.6.mlp.experts.125.down_proj", "model.layers.6.mlp.experts.125.gate_proj", "model.layers.6.mlp.experts.125.up_proj", "model.layers.6.mlp.experts.126.down_proj", "model.layers.6.mlp.experts.126.gate_proj", "model.layers.6.mlp.experts.126.up_proj", "model.layers.6.mlp.experts.127.down_proj", "model.layers.6.mlp.experts.127.gate_proj", "model.layers.6.mlp.experts.127.up_proj", "model.layers.6.mlp.experts.128.down_proj", "model.layers.6.mlp.experts.128.gate_proj", "model.layers.6.mlp.experts.128.up_proj", "model.layers.6.mlp.experts.129.down_proj", "model.layers.6.mlp.experts.129.gate_proj", "model.layers.6.mlp.experts.129.up_proj", "model.layers.6.mlp.experts.13.down_proj", "model.layers.6.mlp.experts.13.gate_proj", "model.layers.6.mlp.experts.13.up_proj", "model.layers.6.mlp.experts.130.down_proj", "model.layers.6.mlp.experts.130.gate_proj", "model.layers.6.mlp.experts.130.up_proj", "model.layers.6.mlp.experts.131.down_proj", "model.layers.6.mlp.experts.131.gate_proj", "model.layers.6.mlp.experts.131.up_proj", "model.layers.6.mlp.experts.132.down_proj", "model.layers.6.mlp.experts.132.gate_proj", "model.layers.6.mlp.experts.132.up_proj", "model.layers.6.mlp.experts.133.down_proj", "model.layers.6.mlp.experts.133.gate_proj", "model.layers.6.mlp.experts.133.up_proj", "model.layers.6.mlp.experts.134.down_proj", "model.layers.6.mlp.experts.134.gate_proj", "model.layers.6.mlp.experts.134.up_proj", "model.layers.6.mlp.experts.135.down_proj", "model.layers.6.mlp.experts.135.gate_proj", "model.layers.6.mlp.experts.135.up_proj", "model.layers.6.mlp.experts.136.down_proj", "model.layers.6.mlp.experts.136.gate_proj", "model.layers.6.mlp.experts.136.up_proj", "model.layers.6.mlp.experts.137.down_proj", "model.layers.6.mlp.experts.137.gate_proj", "model.layers.6.mlp.experts.137.up_proj", "model.layers.6.mlp.experts.138.down_proj", "model.layers.6.mlp.experts.138.gate_proj", "model.layers.6.mlp.experts.138.up_proj", "model.layers.6.mlp.experts.139.down_proj", "model.layers.6.mlp.experts.139.gate_proj", "model.layers.6.mlp.experts.139.up_proj", "model.layers.6.mlp.experts.14.down_proj", "model.layers.6.mlp.experts.14.gate_proj", "model.layers.6.mlp.experts.14.up_proj", "model.layers.6.mlp.experts.140.down_proj", "model.layers.6.mlp.experts.140.gate_proj", "model.layers.6.mlp.experts.140.up_proj", "model.layers.6.mlp.experts.141.down_proj", "model.layers.6.mlp.experts.141.gate_proj", "model.layers.6.mlp.experts.141.up_proj", "model.layers.6.mlp.experts.142.down_proj", "model.layers.6.mlp.experts.142.gate_proj", "model.layers.6.mlp.experts.142.up_proj", "model.layers.6.mlp.experts.143.down_proj", "model.layers.6.mlp.experts.143.gate_proj", "model.layers.6.mlp.experts.143.up_proj", "model.layers.6.mlp.experts.144.down_proj", "model.layers.6.mlp.experts.144.gate_proj", "model.layers.6.mlp.experts.144.up_proj", "model.layers.6.mlp.experts.145.down_proj", "model.layers.6.mlp.experts.145.gate_proj", "model.layers.6.mlp.experts.145.up_proj", "model.layers.6.mlp.experts.146.down_proj", "model.layers.6.mlp.experts.146.gate_proj", "model.layers.6.mlp.experts.146.up_proj", "model.layers.6.mlp.experts.147.down_proj", "model.layers.6.mlp.experts.147.gate_proj", "model.layers.6.mlp.experts.147.up_proj", "model.layers.6.mlp.experts.148.down_proj", "model.layers.6.mlp.experts.148.gate_proj", "model.layers.6.mlp.experts.148.up_proj", "model.layers.6.mlp.experts.149.down_proj", "model.layers.6.mlp.experts.149.gate_proj", "model.layers.6.mlp.experts.149.up_proj", "model.layers.6.mlp.experts.15.down_proj", "model.layers.6.mlp.experts.15.gate_proj", "model.layers.6.mlp.experts.15.up_proj", "model.layers.6.mlp.experts.150.down_proj", "model.layers.6.mlp.experts.150.gate_proj", "model.layers.6.mlp.experts.150.up_proj", "model.layers.6.mlp.experts.151.down_proj", "model.layers.6.mlp.experts.151.gate_proj", "model.layers.6.mlp.experts.151.up_proj", "model.layers.6.mlp.experts.152.down_proj", "model.layers.6.mlp.experts.152.gate_proj", "model.layers.6.mlp.experts.152.up_proj", "model.layers.6.mlp.experts.153.down_proj", "model.layers.6.mlp.experts.153.gate_proj", "model.layers.6.mlp.experts.153.up_proj", "model.layers.6.mlp.experts.154.down_proj", "model.layers.6.mlp.experts.154.gate_proj", "model.layers.6.mlp.experts.154.up_proj", "model.layers.6.mlp.experts.155.down_proj", "model.layers.6.mlp.experts.155.gate_proj", "model.layers.6.mlp.experts.155.up_proj", "model.layers.6.mlp.experts.156.down_proj", "model.layers.6.mlp.experts.156.gate_proj", "model.layers.6.mlp.experts.156.up_proj", "model.layers.6.mlp.experts.157.down_proj", "model.layers.6.mlp.experts.157.gate_proj", "model.layers.6.mlp.experts.157.up_proj", "model.layers.6.mlp.experts.158.down_proj", "model.layers.6.mlp.experts.158.gate_proj", "model.layers.6.mlp.experts.158.up_proj", "model.layers.6.mlp.experts.159.down_proj", "model.layers.6.mlp.experts.159.gate_proj", "model.layers.6.mlp.experts.159.up_proj", "model.layers.6.mlp.experts.16.down_proj", "model.layers.6.mlp.experts.16.gate_proj", "model.layers.6.mlp.experts.16.up_proj", "model.layers.6.mlp.experts.160.down_proj", "model.layers.6.mlp.experts.160.gate_proj", "model.layers.6.mlp.experts.160.up_proj", "model.layers.6.mlp.experts.161.down_proj", "model.layers.6.mlp.experts.161.gate_proj", "model.layers.6.mlp.experts.161.up_proj", "model.layers.6.mlp.experts.162.down_proj", "model.layers.6.mlp.experts.162.gate_proj", "model.layers.6.mlp.experts.162.up_proj", "model.layers.6.mlp.experts.163.down_proj", "model.layers.6.mlp.experts.163.gate_proj", "model.layers.6.mlp.experts.163.up_proj", "model.layers.6.mlp.experts.164.down_proj", "model.layers.6.mlp.experts.164.gate_proj", "model.layers.6.mlp.experts.164.up_proj", "model.layers.6.mlp.experts.165.down_proj", "model.layers.6.mlp.experts.165.gate_proj", "model.layers.6.mlp.experts.165.up_proj", "model.layers.6.mlp.experts.166.down_proj", "model.layers.6.mlp.experts.166.gate_proj", "model.layers.6.mlp.experts.166.up_proj", "model.layers.6.mlp.experts.167.down_proj", "model.layers.6.mlp.experts.167.gate_proj", "model.layers.6.mlp.experts.167.up_proj", "model.layers.6.mlp.experts.168.down_proj", "model.layers.6.mlp.experts.168.gate_proj", "model.layers.6.mlp.experts.168.up_proj", "model.layers.6.mlp.experts.169.down_proj", "model.layers.6.mlp.experts.169.gate_proj", "model.layers.6.mlp.experts.169.up_proj", "model.layers.6.mlp.experts.17.down_proj", "model.layers.6.mlp.experts.17.gate_proj", "model.layers.6.mlp.experts.17.up_proj", "model.layers.6.mlp.experts.170.down_proj", "model.layers.6.mlp.experts.170.gate_proj", "model.layers.6.mlp.experts.170.up_proj", "model.layers.6.mlp.experts.171.down_proj", "model.layers.6.mlp.experts.171.gate_proj", "model.layers.6.mlp.experts.171.up_proj", "model.layers.6.mlp.experts.172.down_proj", "model.layers.6.mlp.experts.172.gate_proj", "model.layers.6.mlp.experts.172.up_proj", "model.layers.6.mlp.experts.173.down_proj", "model.layers.6.mlp.experts.173.gate_proj", "model.layers.6.mlp.experts.173.up_proj", "model.layers.6.mlp.experts.174.down_proj", "model.layers.6.mlp.experts.174.gate_proj", "model.layers.6.mlp.experts.174.up_proj", "model.layers.6.mlp.experts.175.down_proj", "model.layers.6.mlp.experts.175.gate_proj", "model.layers.6.mlp.experts.175.up_proj", "model.layers.6.mlp.experts.176.down_proj", "model.layers.6.mlp.experts.176.gate_proj", "model.layers.6.mlp.experts.176.up_proj", "model.layers.6.mlp.experts.177.down_proj", "model.layers.6.mlp.experts.177.gate_proj", "model.layers.6.mlp.experts.177.up_proj", "model.layers.6.mlp.experts.178.down_proj", "model.layers.6.mlp.experts.178.gate_proj", "model.layers.6.mlp.experts.178.up_proj", "model.layers.6.mlp.experts.179.down_proj", "model.layers.6.mlp.experts.179.gate_proj", "model.layers.6.mlp.experts.179.up_proj", "model.layers.6.mlp.experts.18.down_proj", "model.layers.6.mlp.experts.18.gate_proj", "model.layers.6.mlp.experts.18.up_proj", "model.layers.6.mlp.experts.180.down_proj", "model.layers.6.mlp.experts.180.gate_proj", "model.layers.6.mlp.experts.180.up_proj", "model.layers.6.mlp.experts.181.down_proj", "model.layers.6.mlp.experts.181.gate_proj", "model.layers.6.mlp.experts.181.up_proj", "model.layers.6.mlp.experts.182.down_proj", "model.layers.6.mlp.experts.182.gate_proj", "model.layers.6.mlp.experts.182.up_proj", "model.layers.6.mlp.experts.183.down_proj", "model.layers.6.mlp.experts.183.gate_proj", "model.layers.6.mlp.experts.183.up_proj", "model.layers.6.mlp.experts.184.down_proj", "model.layers.6.mlp.experts.184.gate_proj", "model.layers.6.mlp.experts.184.up_proj", "model.layers.6.mlp.experts.185.down_proj", "model.layers.6.mlp.experts.185.gate_proj", "model.layers.6.mlp.experts.185.up_proj", "model.layers.6.mlp.experts.186.down_proj", "model.layers.6.mlp.experts.186.gate_proj", "model.layers.6.mlp.experts.186.up_proj", "model.layers.6.mlp.experts.187.down_proj", "model.layers.6.mlp.experts.187.gate_proj", "model.layers.6.mlp.experts.187.up_proj", "model.layers.6.mlp.experts.188.down_proj", "model.layers.6.mlp.experts.188.gate_proj", "model.layers.6.mlp.experts.188.up_proj", "model.layers.6.mlp.experts.189.down_proj", "model.layers.6.mlp.experts.189.gate_proj", "model.layers.6.mlp.experts.189.up_proj", "model.layers.6.mlp.experts.19.down_proj", "model.layers.6.mlp.experts.19.gate_proj", "model.layers.6.mlp.experts.19.up_proj", "model.layers.6.mlp.experts.190.down_proj", "model.layers.6.mlp.experts.190.gate_proj", "model.layers.6.mlp.experts.190.up_proj", "model.layers.6.mlp.experts.191.down_proj", "model.layers.6.mlp.experts.191.gate_proj", "model.layers.6.mlp.experts.191.up_proj", "model.layers.6.mlp.experts.192.down_proj", "model.layers.6.mlp.experts.192.gate_proj", "model.layers.6.mlp.experts.192.up_proj", "model.layers.6.mlp.experts.193.down_proj", "model.layers.6.mlp.experts.193.gate_proj", "model.layers.6.mlp.experts.193.up_proj", "model.layers.6.mlp.experts.194.down_proj", "model.layers.6.mlp.experts.194.gate_proj", "model.layers.6.mlp.experts.194.up_proj", "model.layers.6.mlp.experts.195.down_proj", "model.layers.6.mlp.experts.195.gate_proj", "model.layers.6.mlp.experts.195.up_proj", "model.layers.6.mlp.experts.196.down_proj", "model.layers.6.mlp.experts.196.gate_proj", "model.layers.6.mlp.experts.196.up_proj", "model.layers.6.mlp.experts.197.down_proj", "model.layers.6.mlp.experts.197.gate_proj", "model.layers.6.mlp.experts.197.up_proj", "model.layers.6.mlp.experts.198.down_proj", "model.layers.6.mlp.experts.198.gate_proj", "model.layers.6.mlp.experts.198.up_proj", "model.layers.6.mlp.experts.199.down_proj", "model.layers.6.mlp.experts.199.gate_proj", "model.layers.6.mlp.experts.199.up_proj", "model.layers.6.mlp.experts.2.down_proj", "model.layers.6.mlp.experts.2.gate_proj", "model.layers.6.mlp.experts.2.up_proj", "model.layers.6.mlp.experts.20.down_proj", "model.layers.6.mlp.experts.20.gate_proj", "model.layers.6.mlp.experts.20.up_proj", "model.layers.6.mlp.experts.200.down_proj", "model.layers.6.mlp.experts.200.gate_proj", "model.layers.6.mlp.experts.200.up_proj", "model.layers.6.mlp.experts.201.down_proj", "model.layers.6.mlp.experts.201.gate_proj", "model.layers.6.mlp.experts.201.up_proj", "model.layers.6.mlp.experts.202.down_proj", "model.layers.6.mlp.experts.202.gate_proj", "model.layers.6.mlp.experts.202.up_proj", "model.layers.6.mlp.experts.203.down_proj", "model.layers.6.mlp.experts.203.gate_proj", "model.layers.6.mlp.experts.203.up_proj", "model.layers.6.mlp.experts.204.down_proj", "model.layers.6.mlp.experts.204.gate_proj", "model.layers.6.mlp.experts.204.up_proj", "model.layers.6.mlp.experts.205.down_proj", "model.layers.6.mlp.experts.205.gate_proj", "model.layers.6.mlp.experts.205.up_proj", "model.layers.6.mlp.experts.206.down_proj", "model.layers.6.mlp.experts.206.gate_proj", "model.layers.6.mlp.experts.206.up_proj", "model.layers.6.mlp.experts.207.down_proj", "model.layers.6.mlp.experts.207.gate_proj", "model.layers.6.mlp.experts.207.up_proj", "model.layers.6.mlp.experts.208.down_proj", "model.layers.6.mlp.experts.208.gate_proj", "model.layers.6.mlp.experts.208.up_proj", "model.layers.6.mlp.experts.209.down_proj", "model.layers.6.mlp.experts.209.gate_proj", "model.layers.6.mlp.experts.209.up_proj", "model.layers.6.mlp.experts.21.down_proj", "model.layers.6.mlp.experts.21.gate_proj", "model.layers.6.mlp.experts.21.up_proj", "model.layers.6.mlp.experts.210.down_proj", "model.layers.6.mlp.experts.210.gate_proj", "model.layers.6.mlp.experts.210.up_proj", "model.layers.6.mlp.experts.211.down_proj", "model.layers.6.mlp.experts.211.gate_proj", "model.layers.6.mlp.experts.211.up_proj", "model.layers.6.mlp.experts.212.down_proj", "model.layers.6.mlp.experts.212.gate_proj", "model.layers.6.mlp.experts.212.up_proj", "model.layers.6.mlp.experts.213.down_proj", "model.layers.6.mlp.experts.213.gate_proj", "model.layers.6.mlp.experts.213.up_proj", "model.layers.6.mlp.experts.214.down_proj", "model.layers.6.mlp.experts.214.gate_proj", "model.layers.6.mlp.experts.214.up_proj", "model.layers.6.mlp.experts.215.down_proj", "model.layers.6.mlp.experts.215.gate_proj", "model.layers.6.mlp.experts.215.up_proj", "model.layers.6.mlp.experts.216.down_proj", "model.layers.6.mlp.experts.216.gate_proj", "model.layers.6.mlp.experts.216.up_proj", "model.layers.6.mlp.experts.217.down_proj", "model.layers.6.mlp.experts.217.gate_proj", "model.layers.6.mlp.experts.217.up_proj", "model.layers.6.mlp.experts.218.down_proj", "model.layers.6.mlp.experts.218.gate_proj", "model.layers.6.mlp.experts.218.up_proj", "model.layers.6.mlp.experts.219.down_proj", "model.layers.6.mlp.experts.219.gate_proj", "model.layers.6.mlp.experts.219.up_proj", "model.layers.6.mlp.experts.22.down_proj", "model.layers.6.mlp.experts.22.gate_proj", "model.layers.6.mlp.experts.22.up_proj", "model.layers.6.mlp.experts.220.down_proj", "model.layers.6.mlp.experts.220.gate_proj", "model.layers.6.mlp.experts.220.up_proj", "model.layers.6.mlp.experts.221.down_proj", "model.layers.6.mlp.experts.221.gate_proj", "model.layers.6.mlp.experts.221.up_proj", "model.layers.6.mlp.experts.222.down_proj", "model.layers.6.mlp.experts.222.gate_proj", "model.layers.6.mlp.experts.222.up_proj", "model.layers.6.mlp.experts.223.down_proj", "model.layers.6.mlp.experts.223.gate_proj", "model.layers.6.mlp.experts.223.up_proj", "model.layers.6.mlp.experts.224.down_proj", "model.layers.6.mlp.experts.224.gate_proj", "model.layers.6.mlp.experts.224.up_proj", "model.layers.6.mlp.experts.225.down_proj", "model.layers.6.mlp.experts.225.gate_proj", "model.layers.6.mlp.experts.225.up_proj", "model.layers.6.mlp.experts.226.down_proj", "model.layers.6.mlp.experts.226.gate_proj", "model.layers.6.mlp.experts.226.up_proj", "model.layers.6.mlp.experts.227.down_proj", "model.layers.6.mlp.experts.227.gate_proj", "model.layers.6.mlp.experts.227.up_proj", "model.layers.6.mlp.experts.228.down_proj", "model.layers.6.mlp.experts.228.gate_proj", "model.layers.6.mlp.experts.228.up_proj", "model.layers.6.mlp.experts.229.down_proj", "model.layers.6.mlp.experts.229.gate_proj", "model.layers.6.mlp.experts.229.up_proj", "model.layers.6.mlp.experts.23.down_proj", "model.layers.6.mlp.experts.23.gate_proj", "model.layers.6.mlp.experts.23.up_proj", "model.layers.6.mlp.experts.230.down_proj", "model.layers.6.mlp.experts.230.gate_proj", "model.layers.6.mlp.experts.230.up_proj", "model.layers.6.mlp.experts.231.down_proj", "model.layers.6.mlp.experts.231.gate_proj", "model.layers.6.mlp.experts.231.up_proj", "model.layers.6.mlp.experts.232.down_proj", "model.layers.6.mlp.experts.232.gate_proj", "model.layers.6.mlp.experts.232.up_proj", "model.layers.6.mlp.experts.233.down_proj", "model.layers.6.mlp.experts.233.gate_proj", "model.layers.6.mlp.experts.233.up_proj", "model.layers.6.mlp.experts.234.down_proj", "model.layers.6.mlp.experts.234.gate_proj", "model.layers.6.mlp.experts.234.up_proj", "model.layers.6.mlp.experts.235.down_proj", "model.layers.6.mlp.experts.235.gate_proj", "model.layers.6.mlp.experts.235.up_proj", "model.layers.6.mlp.experts.236.down_proj", "model.layers.6.mlp.experts.236.gate_proj", "model.layers.6.mlp.experts.236.up_proj", "model.layers.6.mlp.experts.237.down_proj", "model.layers.6.mlp.experts.237.gate_proj", "model.layers.6.mlp.experts.237.up_proj", "model.layers.6.mlp.experts.238.down_proj", "model.layers.6.mlp.experts.238.gate_proj", "model.layers.6.mlp.experts.238.up_proj", "model.layers.6.mlp.experts.239.down_proj", "model.layers.6.mlp.experts.239.gate_proj", "model.layers.6.mlp.experts.239.up_proj", "model.layers.6.mlp.experts.24.down_proj", "model.layers.6.mlp.experts.24.gate_proj", "model.layers.6.mlp.experts.24.up_proj", "model.layers.6.mlp.experts.240.down_proj", "model.layers.6.mlp.experts.240.gate_proj", "model.layers.6.mlp.experts.240.up_proj", "model.layers.6.mlp.experts.241.down_proj", "model.layers.6.mlp.experts.241.gate_proj", "model.layers.6.mlp.experts.241.up_proj", "model.layers.6.mlp.experts.242.down_proj", "model.layers.6.mlp.experts.242.gate_proj", "model.layers.6.mlp.experts.242.up_proj", "model.layers.6.mlp.experts.243.down_proj", "model.layers.6.mlp.experts.243.gate_proj", "model.layers.6.mlp.experts.243.up_proj", "model.layers.6.mlp.experts.244.down_proj", "model.layers.6.mlp.experts.244.gate_proj", "model.layers.6.mlp.experts.244.up_proj", "model.layers.6.mlp.experts.245.down_proj", "model.layers.6.mlp.experts.245.gate_proj", "model.layers.6.mlp.experts.245.up_proj", "model.layers.6.mlp.experts.246.down_proj", "model.layers.6.mlp.experts.246.gate_proj", "model.layers.6.mlp.experts.246.up_proj", "model.layers.6.mlp.experts.247.down_proj", "model.layers.6.mlp.experts.247.gate_proj", "model.layers.6.mlp.experts.247.up_proj", "model.layers.6.mlp.experts.248.down_proj", "model.layers.6.mlp.experts.248.gate_proj", "model.layers.6.mlp.experts.248.up_proj", "model.layers.6.mlp.experts.249.down_proj", "model.layers.6.mlp.experts.249.gate_proj", "model.layers.6.mlp.experts.249.up_proj", "model.layers.6.mlp.experts.25.down_proj", "model.layers.6.mlp.experts.25.gate_proj", "model.layers.6.mlp.experts.25.up_proj", "model.layers.6.mlp.experts.250.down_proj", "model.layers.6.mlp.experts.250.gate_proj", "model.layers.6.mlp.experts.250.up_proj", "model.layers.6.mlp.experts.251.down_proj", "model.layers.6.mlp.experts.251.gate_proj", "model.layers.6.mlp.experts.251.up_proj", "model.layers.6.mlp.experts.252.down_proj", "model.layers.6.mlp.experts.252.gate_proj", "model.layers.6.mlp.experts.252.up_proj", "model.layers.6.mlp.experts.253.down_proj", "model.layers.6.mlp.experts.253.gate_proj", "model.layers.6.mlp.experts.253.up_proj", "model.layers.6.mlp.experts.254.down_proj", "model.layers.6.mlp.experts.254.gate_proj", "model.layers.6.mlp.experts.254.up_proj", "model.layers.6.mlp.experts.255.down_proj", "model.layers.6.mlp.experts.255.gate_proj", "model.layers.6.mlp.experts.255.up_proj", "model.layers.6.mlp.experts.256.down_proj", "model.layers.6.mlp.experts.256.gate_proj", "model.layers.6.mlp.experts.256.up_proj", "model.layers.6.mlp.experts.257.down_proj", "model.layers.6.mlp.experts.257.gate_proj", "model.layers.6.mlp.experts.257.up_proj", "model.layers.6.mlp.experts.258.down_proj", "model.layers.6.mlp.experts.258.gate_proj", "model.layers.6.mlp.experts.258.up_proj", "model.layers.6.mlp.experts.259.down_proj", "model.layers.6.mlp.experts.259.gate_proj", "model.layers.6.mlp.experts.259.up_proj", "model.layers.6.mlp.experts.26.down_proj", "model.layers.6.mlp.experts.26.gate_proj", "model.layers.6.mlp.experts.26.up_proj", "model.layers.6.mlp.experts.260.down_proj", "model.layers.6.mlp.experts.260.gate_proj", "model.layers.6.mlp.experts.260.up_proj", "model.layers.6.mlp.experts.261.down_proj", "model.layers.6.mlp.experts.261.gate_proj", "model.layers.6.mlp.experts.261.up_proj", "model.layers.6.mlp.experts.262.down_proj", "model.layers.6.mlp.experts.262.gate_proj", "model.layers.6.mlp.experts.262.up_proj", "model.layers.6.mlp.experts.263.down_proj", "model.layers.6.mlp.experts.263.gate_proj", "model.layers.6.mlp.experts.263.up_proj", "model.layers.6.mlp.experts.264.down_proj", "model.layers.6.mlp.experts.264.gate_proj", "model.layers.6.mlp.experts.264.up_proj", "model.layers.6.mlp.experts.265.down_proj", "model.layers.6.mlp.experts.265.gate_proj", "model.layers.6.mlp.experts.265.up_proj", "model.layers.6.mlp.experts.266.down_proj", "model.layers.6.mlp.experts.266.gate_proj", "model.layers.6.mlp.experts.266.up_proj", "model.layers.6.mlp.experts.267.down_proj", "model.layers.6.mlp.experts.267.gate_proj", "model.layers.6.mlp.experts.267.up_proj", "model.layers.6.mlp.experts.268.down_proj", "model.layers.6.mlp.experts.268.gate_proj", "model.layers.6.mlp.experts.268.up_proj", "model.layers.6.mlp.experts.269.down_proj", "model.layers.6.mlp.experts.269.gate_proj", "model.layers.6.mlp.experts.269.up_proj", "model.layers.6.mlp.experts.27.down_proj", "model.layers.6.mlp.experts.27.gate_proj", "model.layers.6.mlp.experts.27.up_proj", "model.layers.6.mlp.experts.270.down_proj", "model.layers.6.mlp.experts.270.gate_proj", "model.layers.6.mlp.experts.270.up_proj", "model.layers.6.mlp.experts.271.down_proj", "model.layers.6.mlp.experts.271.gate_proj", "model.layers.6.mlp.experts.271.up_proj", "model.layers.6.mlp.experts.272.down_proj", "model.layers.6.mlp.experts.272.gate_proj", "model.layers.6.mlp.experts.272.up_proj", "model.layers.6.mlp.experts.273.down_proj", "model.layers.6.mlp.experts.273.gate_proj", "model.layers.6.mlp.experts.273.up_proj", "model.layers.6.mlp.experts.274.down_proj", "model.layers.6.mlp.experts.274.gate_proj", "model.layers.6.mlp.experts.274.up_proj", "model.layers.6.mlp.experts.275.down_proj", "model.layers.6.mlp.experts.275.gate_proj", "model.layers.6.mlp.experts.275.up_proj", "model.layers.6.mlp.experts.276.down_proj", "model.layers.6.mlp.experts.276.gate_proj", "model.layers.6.mlp.experts.276.up_proj", "model.layers.6.mlp.experts.277.down_proj", "model.layers.6.mlp.experts.277.gate_proj", "model.layers.6.mlp.experts.277.up_proj", "model.layers.6.mlp.experts.278.down_proj", "model.layers.6.mlp.experts.278.gate_proj", "model.layers.6.mlp.experts.278.up_proj", "model.layers.6.mlp.experts.279.down_proj", "model.layers.6.mlp.experts.279.gate_proj", "model.layers.6.mlp.experts.279.up_proj", "model.layers.6.mlp.experts.28.down_proj", "model.layers.6.mlp.experts.28.gate_proj", "model.layers.6.mlp.experts.28.up_proj", "model.layers.6.mlp.experts.280.down_proj", "model.layers.6.mlp.experts.280.gate_proj", "model.layers.6.mlp.experts.280.up_proj", "model.layers.6.mlp.experts.281.down_proj", "model.layers.6.mlp.experts.281.gate_proj", "model.layers.6.mlp.experts.281.up_proj", "model.layers.6.mlp.experts.282.down_proj", "model.layers.6.mlp.experts.282.gate_proj", "model.layers.6.mlp.experts.282.up_proj", "model.layers.6.mlp.experts.283.down_proj", "model.layers.6.mlp.experts.283.gate_proj", "model.layers.6.mlp.experts.283.up_proj", "model.layers.6.mlp.experts.284.down_proj", "model.layers.6.mlp.experts.284.gate_proj", "model.layers.6.mlp.experts.284.up_proj", "model.layers.6.mlp.experts.285.down_proj", "model.layers.6.mlp.experts.285.gate_proj", "model.layers.6.mlp.experts.285.up_proj", "model.layers.6.mlp.experts.286.down_proj", "model.layers.6.mlp.experts.286.gate_proj", "model.layers.6.mlp.experts.286.up_proj", "model.layers.6.mlp.experts.287.down_proj", "model.layers.6.mlp.experts.287.gate_proj", "model.layers.6.mlp.experts.287.up_proj", "model.layers.6.mlp.experts.29.down_proj", "model.layers.6.mlp.experts.29.gate_proj", "model.layers.6.mlp.experts.29.up_proj", "model.layers.6.mlp.experts.3.down_proj", "model.layers.6.mlp.experts.3.gate_proj", "model.layers.6.mlp.experts.3.up_proj", "model.layers.6.mlp.experts.30.down_proj", "model.layers.6.mlp.experts.30.gate_proj", "model.layers.6.mlp.experts.30.up_proj", "model.layers.6.mlp.experts.31.down_proj", "model.layers.6.mlp.experts.31.gate_proj", "model.layers.6.mlp.experts.31.up_proj", "model.layers.6.mlp.experts.32.down_proj", "model.layers.6.mlp.experts.32.gate_proj", "model.layers.6.mlp.experts.32.up_proj", "model.layers.6.mlp.experts.33.down_proj", "model.layers.6.mlp.experts.33.gate_proj", "model.layers.6.mlp.experts.33.up_proj", "model.layers.6.mlp.experts.34.down_proj", "model.layers.6.mlp.experts.34.gate_proj", "model.layers.6.mlp.experts.34.up_proj", "model.layers.6.mlp.experts.35.down_proj", "model.layers.6.mlp.experts.35.gate_proj", "model.layers.6.mlp.experts.35.up_proj", "model.layers.6.mlp.experts.36.down_proj", "model.layers.6.mlp.experts.36.gate_proj", "model.layers.6.mlp.experts.36.up_proj", "model.layers.6.mlp.experts.37.down_proj", "model.layers.6.mlp.experts.37.gate_proj", "model.layers.6.mlp.experts.37.up_proj", "model.layers.6.mlp.experts.38.down_proj", "model.layers.6.mlp.experts.38.gate_proj", "model.layers.6.mlp.experts.38.up_proj", "model.layers.6.mlp.experts.39.down_proj", "model.layers.6.mlp.experts.39.gate_proj", "model.layers.6.mlp.experts.39.up_proj", "model.layers.6.mlp.experts.4.down_proj", "model.layers.6.mlp.experts.4.gate_proj", "model.layers.6.mlp.experts.4.up_proj", "model.layers.6.mlp.experts.40.down_proj", "model.layers.6.mlp.experts.40.gate_proj", "model.layers.6.mlp.experts.40.up_proj", "model.layers.6.mlp.experts.41.down_proj", "model.layers.6.mlp.experts.41.gate_proj", "model.layers.6.mlp.experts.41.up_proj", "model.layers.6.mlp.experts.42.down_proj", "model.layers.6.mlp.experts.42.gate_proj", "model.layers.6.mlp.experts.42.up_proj", "model.layers.6.mlp.experts.43.down_proj", "model.layers.6.mlp.experts.43.gate_proj", "model.layers.6.mlp.experts.43.up_proj", "model.layers.6.mlp.experts.44.down_proj", "model.layers.6.mlp.experts.44.gate_proj", "model.layers.6.mlp.experts.44.up_proj", "model.layers.6.mlp.experts.45.down_proj", "model.layers.6.mlp.experts.45.gate_proj", "model.layers.6.mlp.experts.45.up_proj", "model.layers.6.mlp.experts.46.down_proj", "model.layers.6.mlp.experts.46.gate_proj", "model.layers.6.mlp.experts.46.up_proj", "model.layers.6.mlp.experts.47.down_proj", "model.layers.6.mlp.experts.47.gate_proj", "model.layers.6.mlp.experts.47.up_proj", "model.layers.6.mlp.experts.48.down_proj", "model.layers.6.mlp.experts.48.gate_proj", "model.layers.6.mlp.experts.48.up_proj", "model.layers.6.mlp.experts.49.down_proj", "model.layers.6.mlp.experts.49.gate_proj", "model.layers.6.mlp.experts.49.up_proj", "model.layers.6.mlp.experts.5.down_proj", "model.layers.6.mlp.experts.5.gate_proj", "model.layers.6.mlp.experts.5.up_proj", "model.layers.6.mlp.experts.50.down_proj", "model.layers.6.mlp.experts.50.gate_proj", "model.layers.6.mlp.experts.50.up_proj", "model.layers.6.mlp.experts.51.down_proj", "model.layers.6.mlp.experts.51.gate_proj", "model.layers.6.mlp.experts.51.up_proj", "model.layers.6.mlp.experts.52.down_proj", "model.layers.6.mlp.experts.52.gate_proj", "model.layers.6.mlp.experts.52.up_proj", "model.layers.6.mlp.experts.53.down_proj", "model.layers.6.mlp.experts.53.gate_proj", "model.layers.6.mlp.experts.53.up_proj", "model.layers.6.mlp.experts.54.down_proj", "model.layers.6.mlp.experts.54.gate_proj", "model.layers.6.mlp.experts.54.up_proj", "model.layers.6.mlp.experts.55.down_proj", "model.layers.6.mlp.experts.55.gate_proj", "model.layers.6.mlp.experts.55.up_proj", "model.layers.6.mlp.experts.56.down_proj", "model.layers.6.mlp.experts.56.gate_proj", "model.layers.6.mlp.experts.56.up_proj", "model.layers.6.mlp.experts.57.down_proj", "model.layers.6.mlp.experts.57.gate_proj", "model.layers.6.mlp.experts.57.up_proj", "model.layers.6.mlp.experts.58.down_proj", "model.layers.6.mlp.experts.58.gate_proj", "model.layers.6.mlp.experts.58.up_proj", "model.layers.6.mlp.experts.59.down_proj", "model.layers.6.mlp.experts.59.gate_proj", "model.layers.6.mlp.experts.59.up_proj", "model.layers.6.mlp.experts.6.down_proj", "model.layers.6.mlp.experts.6.gate_proj", "model.layers.6.mlp.experts.6.up_proj", "model.layers.6.mlp.experts.60.down_proj", "model.layers.6.mlp.experts.60.gate_proj", "model.layers.6.mlp.experts.60.up_proj", "model.layers.6.mlp.experts.61.down_proj", "model.layers.6.mlp.experts.61.gate_proj", "model.layers.6.mlp.experts.61.up_proj", "model.layers.6.mlp.experts.62.down_proj", "model.layers.6.mlp.experts.62.gate_proj", "model.layers.6.mlp.experts.62.up_proj", "model.layers.6.mlp.experts.63.down_proj", "model.layers.6.mlp.experts.63.gate_proj", "model.layers.6.mlp.experts.63.up_proj", "model.layers.6.mlp.experts.64.down_proj", "model.layers.6.mlp.experts.64.gate_proj", "model.layers.6.mlp.experts.64.up_proj", "model.layers.6.mlp.experts.65.down_proj", "model.layers.6.mlp.experts.65.gate_proj", "model.layers.6.mlp.experts.65.up_proj", "model.layers.6.mlp.experts.66.down_proj", "model.layers.6.mlp.experts.66.gate_proj", "model.layers.6.mlp.experts.66.up_proj", "model.layers.6.mlp.experts.67.down_proj", "model.layers.6.mlp.experts.67.gate_proj", "model.layers.6.mlp.experts.67.up_proj", "model.layers.6.mlp.experts.68.down_proj", "model.layers.6.mlp.experts.68.gate_proj", "model.layers.6.mlp.experts.68.up_proj", "model.layers.6.mlp.experts.69.down_proj", "model.layers.6.mlp.experts.69.gate_proj", "model.layers.6.mlp.experts.69.up_proj", "model.layers.6.mlp.experts.7.down_proj", "model.layers.6.mlp.experts.7.gate_proj", "model.layers.6.mlp.experts.7.up_proj", "model.layers.6.mlp.experts.70.down_proj", "model.layers.6.mlp.experts.70.gate_proj", "model.layers.6.mlp.experts.70.up_proj", "model.layers.6.mlp.experts.71.down_proj", "model.layers.6.mlp.experts.71.gate_proj", "model.layers.6.mlp.experts.71.up_proj", "model.layers.6.mlp.experts.72.down_proj", "model.layers.6.mlp.experts.72.gate_proj", "model.layers.6.mlp.experts.72.up_proj", "model.layers.6.mlp.experts.73.down_proj", "model.layers.6.mlp.experts.73.gate_proj", "model.layers.6.mlp.experts.73.up_proj", "model.layers.6.mlp.experts.74.down_proj", "model.layers.6.mlp.experts.74.gate_proj", "model.layers.6.mlp.experts.74.up_proj", "model.layers.6.mlp.experts.75.down_proj", "model.layers.6.mlp.experts.75.gate_proj", "model.layers.6.mlp.experts.75.up_proj", "model.layers.6.mlp.experts.76.down_proj", "model.layers.6.mlp.experts.76.gate_proj", "model.layers.6.mlp.experts.76.up_proj", "model.layers.6.mlp.experts.77.down_proj", "model.layers.6.mlp.experts.77.gate_proj", "model.layers.6.mlp.experts.77.up_proj", "model.layers.6.mlp.experts.78.down_proj", "model.layers.6.mlp.experts.78.gate_proj", "model.layers.6.mlp.experts.78.up_proj", "model.layers.6.mlp.experts.79.down_proj", "model.layers.6.mlp.experts.79.gate_proj", "model.layers.6.mlp.experts.79.up_proj", "model.layers.6.mlp.experts.8.down_proj", "model.layers.6.mlp.experts.8.gate_proj", "model.layers.6.mlp.experts.8.up_proj", "model.layers.6.mlp.experts.80.down_proj", "model.layers.6.mlp.experts.80.gate_proj", "model.layers.6.mlp.experts.80.up_proj", "model.layers.6.mlp.experts.81.down_proj", "model.layers.6.mlp.experts.81.gate_proj", "model.layers.6.mlp.experts.81.up_proj", "model.layers.6.mlp.experts.82.down_proj", "model.layers.6.mlp.experts.82.gate_proj", "model.layers.6.mlp.experts.82.up_proj", "model.layers.6.mlp.experts.83.down_proj", "model.layers.6.mlp.experts.83.gate_proj", "model.layers.6.mlp.experts.83.up_proj", "model.layers.6.mlp.experts.84.down_proj", "model.layers.6.mlp.experts.84.gate_proj", "model.layers.6.mlp.experts.84.up_proj", "model.layers.6.mlp.experts.85.down_proj", "model.layers.6.mlp.experts.85.gate_proj", "model.layers.6.mlp.experts.85.up_proj", "model.layers.6.mlp.experts.86.down_proj", "model.layers.6.mlp.experts.86.gate_proj", "model.layers.6.mlp.experts.86.up_proj", "model.layers.6.mlp.experts.87.down_proj", "model.layers.6.mlp.experts.87.gate_proj", "model.layers.6.mlp.experts.87.up_proj", "model.layers.6.mlp.experts.88.down_proj", "model.layers.6.mlp.experts.88.gate_proj", "model.layers.6.mlp.experts.88.up_proj", "model.layers.6.mlp.experts.89.down_proj", "model.layers.6.mlp.experts.89.gate_proj", "model.layers.6.mlp.experts.89.up_proj", "model.layers.6.mlp.experts.9.down_proj", "model.layers.6.mlp.experts.9.gate_proj", "model.layers.6.mlp.experts.9.up_proj", "model.layers.6.mlp.experts.90.down_proj", "model.layers.6.mlp.experts.90.gate_proj", "model.layers.6.mlp.experts.90.up_proj", "model.layers.6.mlp.experts.91.down_proj", "model.layers.6.mlp.experts.91.gate_proj", "model.layers.6.mlp.experts.91.up_proj", "model.layers.6.mlp.experts.92.down_proj", "model.layers.6.mlp.experts.92.gate_proj", "model.layers.6.mlp.experts.92.up_proj", "model.layers.6.mlp.experts.93.down_proj", "model.layers.6.mlp.experts.93.gate_proj", "model.layers.6.mlp.experts.93.up_proj", "model.layers.6.mlp.experts.94.down_proj", "model.layers.6.mlp.experts.94.gate_proj", "model.layers.6.mlp.experts.94.up_proj", "model.layers.6.mlp.experts.95.down_proj", "model.layers.6.mlp.experts.95.gate_proj", "model.layers.6.mlp.experts.95.up_proj", "model.layers.6.mlp.experts.96.down_proj", "model.layers.6.mlp.experts.96.gate_proj", "model.layers.6.mlp.experts.96.up_proj", "model.layers.6.mlp.experts.97.down_proj", "model.layers.6.mlp.experts.97.gate_proj", "model.layers.6.mlp.experts.97.up_proj", "model.layers.6.mlp.experts.98.down_proj", "model.layers.6.mlp.experts.98.gate_proj", "model.layers.6.mlp.experts.98.up_proj", "model.layers.6.mlp.experts.99.down_proj", "model.layers.6.mlp.experts.99.gate_proj", "model.layers.6.mlp.experts.99.up_proj" ], "export": { "kv_cache_group": [], "min_kv_scale": 0.0, "pack_method": "reorder", "weight_format": "real_quantized", "weight_merge_groups": null }, "global_quant_config": { "bias": null, "input_tensors": { "block_size": null, "ch_axis": -1, "dtype": "fp4", "enable_buffer_reuse": false, "group_size": 32, "is_dynamic": true, "is_scale_quant": false, "max_input_numel": 4194304, "mx_element_dtype": null, "observer_cls": "PerBlockMXObserver", "qscheme": "per_group", "round_method": "half_even", "scale_calculation_mode": "even", "scale_format": "e8m0", "scale_type": "float", "symmetric": null }, "output_tensors": null, "target_device": null, "weight": { "block_size": null, "ch_axis": -1, "dtype": "fp4", "enable_buffer_reuse": false, "group_size": 32, "is_dynamic": false, "is_scale_quant": false, "max_input_numel": 4194304, "mx_element_dtype": null, "observer_cls": "PerBlockMXObserver", "qscheme": "per_group", "round_method": "half_even", "scale_calculation_mode": "even", "scale_format": "e8m0", "scale_type": "float", "symmetric": null } }, "kv_cache_post_rope": false, "kv_cache_quant_config": {}, "layer_quant_config": {}, "layer_type_quant_config": {}, "quant_method": "quark", "quant_mode": "eager_mode", "softmax_quant_spec": null, "version": "0.12.post1" }, "text_config": { "attention_bias": false, "attention_dropout": 0.0, "dtype": "bfloat16", "eos_token_id": [ 154820, 154827, 154829 ], "first_k_dense_replace": 3, "hc_eps": 1e-06, "hc_mult": 4, "hc_sinkhorn_iters": 20, "head_dim": 0, "hidden_act": "silu", "hidden_size": 4096, "index_head_dim": 128, "index_kpool": 4, "index_kpool_always_select_tail": true, "index_kpool_compress": true, "index_n_heads": 32, "index_share_for_mtp_iteration": true, "index_topk": 2048, "indexer_rope_interleave": true, "indexer_types": [ "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full" ], "initializer_range": 0.02, "intermediate_size": 12288, "kv_lora_rank": 512, "layer_types": [ "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention" ], "linear_attn_config": { "full_attn_layers": [ 3, 7, 11, 15, 19, 23, 27, 31, 35, 39, 43 ], "gate_lower_bound": -5.0, "head_dim": 128, "kda_layers": [ 0, 1, 2, 4, 5, 6, 8, 9, 10, 12, 13, 14, 16, 17, 18, 20, 21, 22, 24, 25, 26, 28, 29, 30, 32, 33, 34, 36, 37, 38, 40, 41, 42, 44 ], "num_heads": 64, "short_conv_kernel_size": 4 }, "max_position_embeddings": 1048576, "mhc": true, "mla_use_nope": true, "mlp_layer_types": [ "dense", "dense", "dense", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse" ], "model_type": "glm5_next_text", "moe_intermediate_size": 2048, "moe_router_dtype": "float32", "n_group": 1, "n_routed_experts": 288, "n_shared_experts": 1, "norm_topk_prob": true, "num_attention_heads": 64, "num_experts_per_tok": 8, "num_hidden_layers": 45, "num_key_value_heads": 64, "num_nextn_predict_layers": 1, "output_router_logits": false, "pad_token_id": 154820, "q_lora_rank": 1536, "qk_head_dim": 256, "qk_nope_head_dim": 256, "qk_rope_head_dim": 0, "rms_norm_eps": 1e-05, "routed_scaling_factor": 2.5, "router_aux_loss_coef": 0.001, "scoring_func": "sigmoid", "swiglu_limit": 10.0, "tie_word_embeddings": false, "topk_group": 1, "topk_method": "noaux_tc", "use_cache": true, "v_head_dim": 256, "vocab_size": 154880 }, "tie_word_embeddings": false, "transformers_version": "5.16.0", "video_end_token_id": 154833, "video_start_token_id": 154832, "video_token_id": 154855, "vision_config": { "attention_bias": true, "attention_dropout": 0.0, "depth": 24, "hidden_act": "silu", "hidden_size": 1024, "image_size": 448, "in_channels": 3, "initializer_range": 0.02, "intermediate_size": 4096, "model_type": "glm5_next_vision", "num_heads": 16, "out_hidden_size": 4096, "patch_size": 14, "projection_intermediate_size": 10240, "rms_norm_eps": 1e-05, "spatial_merge_size": 2, "swiglu_limit": 10.0, "temporal_patch_size": 2 } }