diff --git "a/model.safetensors.index.json.meta.json" "b/model.safetensors.index.json.meta.json" new file mode 100644--- /dev/null +++ "b/model.safetensors.index.json.meta.json" @@ -0,0 +1 @@ +{"model.layers.0.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.0.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.0.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.0.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.0.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.0.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.0.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.1.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.2.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.3.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.4.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.5.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.6.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.7.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.8.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.9.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.10.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.11.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.12.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.13.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.14.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.15.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.16.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.17.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.18.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.19.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.20.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.21.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.22.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.23.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.24.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.25.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.26.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.27.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.28.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.29.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.30.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.31.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.32.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.33.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.34.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.35.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.36.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.37.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.38.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.39.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.40.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.41.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.42.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.43.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.44.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.45.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.46.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.47.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.48.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.49.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.50.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.51.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.52.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.53.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.54.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.55.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.56.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.57.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.58.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.59.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.60.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.61.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.62.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.self_attn.q_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [8192, 5120], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.self_attn.k_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.self_attn.v_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [1024, 5120], "scale.shape": [81920, 1], "awq_scale": null, "zero.shape": [81920, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.self_attn.o_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 8192], "scale.shape": [655360, 1], "awq_scale": null, "zero.shape": [655360, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.mlp.gate_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.mlp.up_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [25600, 5120], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}, "model.layers.63.mlp.down_proj": {"meta": {"nbits": 3, "group_size": 64, "shape": [5120, 25600], "scale.shape": [2048000, 1], "awq_scale": null, "zero.shape": [2048000, 1], "axis": 1, "packing": "3bit_32", "method": "sinq_awq_l1_quantAux", "compute_dtype": "torch.bfloat16", "unpack_view_dtype": "torch.int32", "view_as_float": false, "quant_scale": false, "quant_zero": false}}} \ No newline at end of file