| { |
| "bits": 4, |
| "data_type": "int", |
| "group_size": 128, |
| "sym": true, |
| "autoround_version": "0.13.0", |
| "block_name_to_quantize": "model.language_model.layers", |
| "quant_method": "auto-round", |
| "packing_format": "auto_round:auto_gptq", |
| "extra_config": { |
| "model.language_model.layers.0.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.1.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.2.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.3.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.4.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.5.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.6.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.7.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.8.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.9.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.10.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.11.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.12.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.13.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.14.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.15.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.16.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.17.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.18.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.19.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.20.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.21.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.22.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.23.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.24.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.25.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.26.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.27.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.28.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.29.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.30.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.31.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.32.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.33.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.34.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.35.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.36.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.37.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.38.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| "model.language_model.layers.39.mlp.shared_expert_gate": { |
| "bits": 16, |
| "data_type": "fp" |
| }, |
| ".*model\\.language_model\\.layers\\.0\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.1\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.2\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.3\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.4\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.5\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.6\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.7\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.8\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.9\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.10\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.11\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.12\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.13\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.14\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.15\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.16\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.17\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.18\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.19\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.20\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.21\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.22\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.23\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.24\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.25\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.26\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.27\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.28\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.29\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.30\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.31\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.32\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.33\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.34\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.35\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.36\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.37\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.38\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*model\\.language_model\\.layers\\.39\\.mlp\\.gate.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.0.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.0.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.0.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.0.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.0.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.0.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.1.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.1.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.1.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.1.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.1.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.1.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.10.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.10.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.10.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.10.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.10.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.10.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.11.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.12.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.12.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.12.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.12.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.12.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.12.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.13.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.13.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.13.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.13.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.13.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.13.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.14.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.14.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.14.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.14.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.14.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.14.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.15.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.16.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.16.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.16.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.16.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.16.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.16.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.17.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.17.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.17.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.17.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.17.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.17.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.18.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.18.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.18.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.18.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.18.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.18.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.19.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.2.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.2.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.2.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.2.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.2.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.2.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.20.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.20.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.20.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.20.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.20.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.20.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.21.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.21.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.21.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.21.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.21.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.21.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.22.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.22.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.22.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.22.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.22.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.22.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.23.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.24.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.24.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.24.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.24.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.24.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.24.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.25.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.25.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.25.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.25.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.25.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.25.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.26.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.26.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.26.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.26.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.26.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.26.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.27.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.28.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.28.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.28.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.28.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.28.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.28.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.29.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.29.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.29.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.29.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.29.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.29.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.3.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.30.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.30.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.30.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.30.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.30.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.30.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.31.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.32.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.32.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.32.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.32.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.32.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.32.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.33.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.33.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.33.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.33.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.33.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.33.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.34.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.34.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.34.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.34.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.34.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.34.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.35.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.36.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.36.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.36.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.36.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.36.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.36.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.37.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.37.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.37.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.37.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.37.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.37.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.38.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.38.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.38.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.38.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.38.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.38.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.39.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.4.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.4.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.4.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.4.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.4.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.4.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.5.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.5.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.5.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.5.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.5.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.5.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.6.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.6.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.6.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.6.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.6.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.6.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.7.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.8.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.8.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.8.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.8.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.8.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.8.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.9.linear_attn.in_proj_qkv": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.9.linear_attn.in_proj_z": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.9.linear_attn.out_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.9.mlp.shared_expert.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.9.mlp.shared_expert.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.language_model.layers.9.mlp.shared_expert.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.0\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.0\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.0\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.0\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.10\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.10\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.10\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.10\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.11\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.11\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.11\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.11\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.12\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.12\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.12\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.12\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.13\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.13\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.13\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.13\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.14\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.14\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.14\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.14\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.15\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.15\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.15\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.15\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.16\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.16\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.16\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.16\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.17\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.17\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.17\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.17\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.18\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.18\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.18\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.18\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.19\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.19\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.19\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.19\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.1\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.1\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.1\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.1\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.20\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.20\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.20\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.20\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.21\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.21\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.21\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.21\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.22\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.22\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.22\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.22\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.23\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.23\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.23\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.23\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.24\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.24\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.24\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.24\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.25\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.25\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.25\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.25\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.26\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.26\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.26\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.26\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.27\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.27\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.27\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.27\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.28\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.28\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.28\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.28\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.29\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.29\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.29\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.29\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.2\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.2\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.2\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.2\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.30\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.30\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.30\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.30\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.31\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.31\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.31\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.31\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.32\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.32\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.32\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.32\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.33\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.33\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.33\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.33\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.34\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.34\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.34\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.34\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.35\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.35\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.35\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.35\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.36\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.36\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.36\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.36\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.37\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.37\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.37\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.37\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.38\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.38\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.38\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.38\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.39\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.39\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.39\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.39\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.3\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.3\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.3\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.3\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.4\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.4\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.4\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.4\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.5\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.5\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.5\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.5\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.6\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.6\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.6\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.6\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.7\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.7\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.7\\.self_attn\\.(?:q_proj|k_proj|v_proj|qkv_proj|qkqkv_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.7\\.self_attn\\.o_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.8\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.8\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.8\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.8\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.9\\.linear_attn\\.(?:in_proj_qkv|in_proj_z|in_proj_qkvz)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.9\\.linear_attn\\.out_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.9\\.mlp\\.shared_expert\\.(?:gate_proj|up_proj|gate_up_proj)$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| }, |
| "^(?:model\\.)?(?:language_model\\.)?layers\\.9\\.mlp\\.shared_expert\\.down_proj$": { |
| "bits": 16, |
| "group_size": -1, |
| "sym": true |
| } |
| } |
| } |
|
|
