| { |
| "bits": 4, |
| "data_type": "int", |
| "group_size": 128, |
| "sym": true, |
| "iters": 0, |
| "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.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.self_attn.v_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.self_attn.k_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.self_attn.q_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.self_attn.o_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.mlp.gate_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.mlp.up_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.mlp.down_proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.0.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.1.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.2.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.3.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.4.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.5.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.6.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.7.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.8.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.9.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.10.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.11.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.12.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.13.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.14.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.15.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.16.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.17.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.18.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.19.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.20.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.21.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.22.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.23.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.24.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.25.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.26.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.27.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.28.router.proj": { |
| "bits": 8 |
| }, |
| "model.language_model.layers.29.router.proj": { |
| "bits": 8 |
| }, |
| ".*model\\.language_model\\.layers\\.\\d+\\.self_attn\\..*": { |
| "bits": 8 |
| }, |
| ".*model\\.language_model\\.layers\\.\\d+\\.mlp\\..*": { |
| "bits": 8 |
| }, |
| ".*model\\.language_model\\.layers\\.\\d+\\.router\\..*": { |
| "bits": 8 |
| } |
| } |
| } |