{ "bits": 8, "act_bits": 8, "data_type": "mx_fp", "act_data_type": "mx_fp", "group_size": 32, "act_group_size": 32, "sym": true, "act_sym": true, "act_dynamic": true, "enable_quanted_input": false, "iters": 50, "static_attention_granularity": "tensor", "static_kv_granularity": "tensor", "autoround_version": "0.16.0", "block_name_to_quantize": [ "model.layers.0", "model.layers.1", "model.layers.2", "model.layers.3", "model.layers.4", "model.layers.5", "model.layers.6", "model.layers.7", "model.layers.8", "model.layers.9", "model.layers.10", "model.layers.11", "model.layers.12", "model.layers.13", "model.layers.14", "model.layers.15", "model.layers.16", "model.layers.17", "model.layers.18", "model.layers.19", "model.layers.20", "model.layers.21", "model.layers.22", "model.layers.23", "model.layers.24", "model.layers.25", "model.layers.26", "model.layers.27", "model.layers.28", "model.layers.29", "model.layers.30", "model.layers.31" ], "quant_method": "auto-round", "packing_format": "auto_round:llm_compressor", "extra_config": { "model.layers.0.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.1.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.2.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.3.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.4.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.5.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.6.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.7.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.8.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.9.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.10.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.11.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.12.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.13.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.14.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.15.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.16.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.17.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.18.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.19.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.20.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.21.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.22.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.23.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.24.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.25.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.26.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.27.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.28.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.29.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.30.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.31.mlp.gate.wg": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.layers.0.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.0.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.1.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.2.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.3.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.4.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.5.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.6.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.7.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.8.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.9.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.10.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.11.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.12.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.13.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.14.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.15.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.16.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.17.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.18.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.19.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.20.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.21.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.22.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.23.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.24.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.25.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.26.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.27.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.28.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.29.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.30.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.0.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.0.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.1.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.1.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.2.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.2.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.3.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.3.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.4.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.4.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.5.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.5.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.6.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.6.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.7.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.7.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.8.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.8.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.9.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.9.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.10.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.10.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.11.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.11.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.12.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.12.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.13.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.13.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.14.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.14.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.15.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.15.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.16.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.16.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.17.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.17.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.18.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.18.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.19.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.19.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.20.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.20.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.21.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.21.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.22.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.22.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.23.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.23.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.24.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.24.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.25.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.25.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.26.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.26.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.27.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.27.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.28.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.28.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.29.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.29.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.30.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.30.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.31.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.31.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.32.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.32.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.33.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.33.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.34.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.34.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.35.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.35.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.36.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.36.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.37.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.37.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.38.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.38.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.39.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.39.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.40.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.40.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.41.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.41.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.42.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.42.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.43.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.43.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.44.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.44.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.45.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.45.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.46.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.46.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.47.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.47.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.48.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.48.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.49.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.49.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.50.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.50.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.51.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.51.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.52.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.52.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.53.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.53.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.54.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.54.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.55.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.55.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.56.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.56.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.57.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.57.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.58.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.58.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.59.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.59.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.60.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.60.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.61.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.61.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.62.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.62.down_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.63.gate_and_up_proj": { "bits": 4, "act_bits": 4 }, "model.layers.31.mlp.experts.63.down_proj": { "bits": 4, "act_bits": 4 }, ".*mlp\\.experts.*": { "bits": 4, "act_bits": 4 }, ".*vae.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*vit.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*vit_aligner.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.0\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.1\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.2\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.3\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.4\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.5\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.6\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.7\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.8\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.9\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.10\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.11\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.12\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.13\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.14\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.15\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.16\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.17\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.18\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.19\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.20\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.21\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.22\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.23\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.24\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.25\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.26\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.27\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.28\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.29\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.30\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, ".*model\\.layers\\.31\\.mlp\\.gate.*": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" } } }