{ "format": "mxfp4", "codebook": "fp4-e2m1", "block_size": 32, "num_modules": 496, "num_weights": 24350556160, "elapsed_sec": 1.464, "quantized_modules": [ { "name": "model.language_model.layers.0.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.340129640288069e-06, "max_abs_err": 0.1796875 }, { "name": "model.language_model.layers.0.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.327000629520626e-06, "max_abs_err": 0.033203125 }, { "name": "model.language_model.layers.0.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.47184038926207e-06, "max_abs_err": 0.014892578125 }, { "name": "model.language_model.layers.0.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 9.302879107053741e-07, "max_abs_err": 0.007080078125 }, { "name": "model.language_model.layers.0.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.022647772217169e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.0.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 9.609526614440256e-07, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.0.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 8.922482379603025e-07, "max_abs_err": 0.01025390625 }, { "name": "model.language_model.layers.0.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.0582490403976408e-06, "max_abs_err": 0.078125 }, { "name": "model.language_model.layers.1.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.4364921955566388e-06, "max_abs_err": 0.1875 }, { "name": "model.language_model.layers.1.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.243701828774647e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.1.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.2719643766322406e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.1.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.199010163574712e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.1.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 5.702012913388899e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.1.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0614443226586445e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.1.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.012209281725518e-06, "max_abs_err": 0.0107421875 }, { "name": "model.language_model.layers.1.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.0954300933008199e-06, "max_abs_err": 0.0400390625 }, { "name": "model.language_model.layers.2.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.482392119418364e-06, "max_abs_err": 0.203125 }, { "name": "model.language_model.layers.2.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.324784190932405e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.2.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.341977051401045e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.2.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.4855718240578426e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.2.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 5.4361921684176195e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.2.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1266802175669e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.2.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0527040785746067e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.2.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1280814078418189e-06, "max_abs_err": 0.068359375 }, { "name": "model.language_model.layers.3.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.790850430756109e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.3.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.373629058638471e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.3.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.5806643861869816e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.3.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3955924461915856e-06, "max_abs_err": 0.19140625 }, { "name": "model.language_model.layers.3.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1323994613121613e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.3.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.090669570658065e-06, "max_abs_err": 0.0107421875 }, { "name": "model.language_model.layers.3.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1240135791013017e-06, "max_abs_err": 0.028076171875 }, { "name": "model.language_model.layers.4.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.4902237782953307e-06, "max_abs_err": 0.1796875 }, { "name": "model.language_model.layers.4.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.405255372650572e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.4.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.543538585086935e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.4.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.01823126594536e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.4.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 6.04235674472875e-06, "max_abs_err": 0.02587890625 }, { "name": "model.language_model.layers.4.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1494702221170883e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.4.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1148127896376536e-06, "max_abs_err": 0.010498046875 }, { "name": "model.language_model.layers.4.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1258000540692592e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.5.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.432014525766135e-06, "max_abs_err": 0.16796875 }, { "name": "model.language_model.layers.5.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3090233298717067e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.5.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.5010533590830164e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.5.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.346092626088648e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.5.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.630851435649674e-06, "max_abs_err": 0.0263671875 }, { "name": "model.language_model.layers.5.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1702524034262751e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.5.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0969224604195915e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.5.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1169224762852537e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.6.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.422219495201716e-06, "max_abs_err": 0.109375 }, { "name": "model.language_model.layers.6.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.398669039394008e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.6.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3648883598070825e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.6.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.4901257802412147e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.6.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.710149823949905e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.6.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1690613064274658e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.6.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0885580650210613e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.6.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.0991014960382017e-06, "max_abs_err": 0.0302734375 }, { "name": "model.language_model.layers.7.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.523200464565889e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.7.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.2130950583232334e-06, "max_abs_err": 0.0263671875 }, { "name": "model.language_model.layers.7.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.3188331397250295e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.7.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2560877823707415e-06, "max_abs_err": 0.11328125 }, { "name": "model.language_model.layers.7.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1942279343202244e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.7.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1067271543652168e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.7.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1184381492057582e-06, "max_abs_err": 0.0224609375 }, { "name": "model.language_model.layers.8.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3926165795273846e-06, "max_abs_err": 0.125 }, { "name": "model.language_model.layers.8.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.328913751625805e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.8.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.457240270814509e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.8.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.414724576738081e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.8.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.7616810004692525e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.8.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1940143167521455e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.8.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1159327186760493e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.8.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1312693004583707e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.9.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3968814275576733e-06, "max_abs_err": 0.10546875 }, { "name": "model.language_model.layers.9.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.323531816728064e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.9.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3604300167789916e-06, "max_abs_err": 0.014404296875 }, { "name": "model.language_model.layers.9.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.7944677185587352e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.9.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.353680196800269e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.9.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1794378451668308e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.9.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1118797829112737e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.9.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1223557976336451e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.10.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.369460389672895e-06, "max_abs_err": 0.14453125 }, { "name": "model.language_model.layers.10.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3439918095391477e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.10.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3184995825431542e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.10.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.390327836110373e-06, "max_abs_err": 0.013916015625 }, { "name": "model.language_model.layers.10.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.115636784263188e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.10.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1490136557767983e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.10.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0944138466584263e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.10.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.0966848549287533e-06, "max_abs_err": 0.012939453125 }, { "name": "model.language_model.layers.11.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.479662043697317e-06, "max_abs_err": 0.04296875 }, { "name": "model.language_model.layers.11.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.291296596013126e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.11.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.354191792619531e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.11.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.228372750323615e-06, "max_abs_err": 0.111328125 }, { "name": "model.language_model.layers.11.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.165185267382185e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.11.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1100120218543452e-06, "max_abs_err": 0.011962890625 }, { "name": "model.language_model.layers.11.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1107923683084664e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.12.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.374474433963769e-06, "max_abs_err": 0.11328125 }, { "name": "model.language_model.layers.12.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3843547296564793e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.12.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3469481220672606e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.12.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.342039124414441e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.12.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.216031356918393e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.12.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1583188097574748e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.12.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1058745030823047e-06, "max_abs_err": 0.009765625 }, { "name": "model.language_model.layers.12.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.10913856588013e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.13.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.396154286543606e-06, "max_abs_err": 0.09375 }, { "name": "model.language_model.layers.13.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4236283024947625e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.13.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.264344175273436e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.13.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.7209880499867722e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.13.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.5268028568680165e-06, "max_abs_err": 0.013916015625 }, { "name": "model.language_model.layers.13.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1495112630655058e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.13.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0930167491096654e-06, "max_abs_err": 0.0107421875 }, { "name": "model.language_model.layers.13.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.0965634373860667e-06, "max_abs_err": 0.0107421875 }, { "name": "model.language_model.layers.14.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3588438580190996e-06, "max_abs_err": 0.109375 }, { "name": "model.language_model.layers.14.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3807181150914403e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.14.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3239497295435285e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.14.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.575647613412002e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.14.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.4657764445000794e-06, "max_abs_err": 0.013916015625 }, { "name": "model.language_model.layers.14.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1364599004082265e-06, "max_abs_err": 0.011962890625 }, { "name": "model.language_model.layers.14.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0844546523003373e-06, "max_abs_err": 0.0087890625 }, { "name": "model.language_model.layers.14.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.086345605472161e-06, "max_abs_err": 0.012451171875 }, { "name": "model.language_model.layers.15.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.360037342441501e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.15.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.3889281237643445e-06, "max_abs_err": 0.0174560546875 }, { "name": "model.language_model.layers.15.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.5406454824405955e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.15.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.213330844824668e-06, "max_abs_err": 0.068359375 }, { "name": "model.language_model.layers.15.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1485496997920563e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.15.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1053830348828342e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.15.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1013668199666427e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.16.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3528089059254853e-06, "max_abs_err": 0.095703125 }, { "name": "model.language_model.layers.16.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4011310415517073e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.16.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.2700576209899737e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.16.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.7853224107966525e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.16.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 4.071529474458657e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.16.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1604623750827159e-06, "max_abs_err": 0.012451171875 }, { "name": "model.language_model.layers.16.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1148498515467509e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.16.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.0999285677826265e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.17.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.369381945754867e-06, "max_abs_err": 0.08203125 }, { "name": "model.language_model.layers.17.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.501376911823172e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.17.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.130911752828979e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.17.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.1590187770925695e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.17.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.5688699426827952e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.17.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.171052872450673e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.17.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1250684792685206e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.17.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1129313861601986e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.18.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.32655384024838e-06, "max_abs_err": 0.109375 }, { "name": "model.language_model.layers.18.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.455300091241952e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.18.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.255201707157539e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.18.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.9159040220984025e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.18.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.522706947478582e-06, "max_abs_err": 0.014404296875 }, { "name": "model.language_model.layers.18.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1864888165291632e-06, "max_abs_err": 0.02587890625 }, { "name": "model.language_model.layers.18.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1311097978250473e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.18.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.112543259296217e-06, "max_abs_err": 0.0400390625 }, { "name": "model.language_model.layers.19.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.3906497972348006e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.19.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.238896286144154e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.19.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.52411405199382e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.19.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3871266421338078e-06, "max_abs_err": 0.12109375 }, { "name": "model.language_model.layers.19.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2077766768925358e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.19.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.137854724220233e-06, "max_abs_err": 0.00830078125 }, { "name": "model.language_model.layers.19.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.142791688835132e-06, "max_abs_err": 0.0546875 }, { "name": "model.language_model.layers.20.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3427837732015178e-06, "max_abs_err": 0.076171875 }, { "name": "model.language_model.layers.20.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.497524292266462e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.20.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.377845248702215e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.20.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.8313673990633106e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.20.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.1266474707081215e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.20.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2283126125112176e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.20.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1292786439298652e-06, "max_abs_err": 0.00927734375 }, { "name": "model.language_model.layers.20.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1417598670959705e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.21.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3215093278849963e-06, "max_abs_err": 0.13671875 }, { "name": "model.language_model.layers.21.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.384746494499268e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.21.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.593970975794946e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.21.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.8469218048267066e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.21.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.691665486054262e-06, "max_abs_err": 0.011962890625 }, { "name": "model.language_model.layers.21.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2315667845541611e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.21.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1398724382161163e-06, "max_abs_err": 0.009033203125 }, { "name": "model.language_model.layers.21.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1464038607300608e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.22.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3195759695227025e-06, "max_abs_err": 0.10546875 }, { "name": "model.language_model.layers.22.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.356267259528977e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.22.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.456388074278948e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.22.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.7984061742026825e-06, "max_abs_err": 0.009033203125 }, { "name": "model.language_model.layers.22.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.6492209599382477e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.22.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2955521242474788e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.22.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1372671906428877e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.22.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1349520718795247e-06, "max_abs_err": 0.060546875 }, { "name": "model.language_model.layers.23.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.2508988877234515e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.23.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.1942678358755074e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.23.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.6610657641867874e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.23.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3509207949246047e-06, "max_abs_err": 0.04296875 }, { "name": "model.language_model.layers.23.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2789765833076672e-06, "max_abs_err": 0.010498046875 }, { "name": "model.language_model.layers.23.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.16176101983001e-06, "max_abs_err": 0.00927734375 }, { "name": "model.language_model.layers.23.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1587710559979314e-06, "max_abs_err": 0.024658203125 }, { "name": "model.language_model.layers.24.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.299099605806987e-06, "max_abs_err": 0.076171875 }, { "name": "model.language_model.layers.24.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3443055852112593e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.24.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.263434680571663e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.24.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.4767548464078573e-06, "max_abs_err": 0.010498046875 }, { "name": "model.language_model.layers.24.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.9717670056706993e-06, "max_abs_err": 0.012939453125 }, { "name": "model.language_model.layers.24.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3004494121560128e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.24.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1739971341739874e-06, "max_abs_err": 0.010009765625 }, { "name": "model.language_model.layers.24.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.169482629848062e-06, "max_abs_err": 0.029541015625 }, { "name": "model.language_model.layers.25.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3556151518278057e-06, "max_abs_err": 0.060546875 }, { "name": "model.language_model.layers.25.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4175160433514975e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.25.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.1943853880657116e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.25.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.3489487855622428e-06, "max_abs_err": 0.010498046875 }, { "name": "model.language_model.layers.25.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.1218193069216795e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.25.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2694434872173588e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.25.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2116912557758042e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.25.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1913406297026086e-06, "max_abs_err": 0.02392578125 }, { "name": "model.language_model.layers.26.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3272611997526838e-06, "max_abs_err": 0.08203125 }, { "name": "model.language_model.layers.26.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4258968096546596e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.26.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.121781562891556e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.26.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.4994369621490478e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.26.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.202995119660045e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.26.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.225907112711866e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.26.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2390913752824417e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.26.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2039535022267955e-06, "max_abs_err": 0.029296875 }, { "name": "model.language_model.layers.27.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.3120792320696637e-06, "max_abs_err": 0.03076171875 }, { "name": "model.language_model.layers.27.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.2842664293420967e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.27.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.4084592951112427e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.27.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2597855604544748e-06, "max_abs_err": 0.04296875 }, { "name": "model.language_model.layers.27.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2046816664224025e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.27.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2543121101771249e-06, "max_abs_err": 0.012451171875 }, { "name": "model.language_model.layers.27.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2168167131676455e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.28.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3123072878661333e-06, "max_abs_err": 0.0703125 }, { "name": "model.language_model.layers.28.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4773764835117618e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.28.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.9960275494668167e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.28.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.4192374919730355e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.28.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.628518359415466e-06, "max_abs_err": 0.012939453125 }, { "name": "model.language_model.layers.28.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1707337534971884e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.28.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2479508768592495e-06, "max_abs_err": 0.0107421875 }, { "name": "model.language_model.layers.28.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2111520391044905e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.29.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3196000711322995e-06, "max_abs_err": 0.064453125 }, { "name": "model.language_model.layers.29.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4906166800064966e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.29.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.946443035194534e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.29.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.52806683217932e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.29.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.0085290088900365e-06, "max_abs_err": 0.010009765625 }, { "name": "model.language_model.layers.29.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1534931445567054e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.29.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2378607152641052e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.29.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1980259841948282e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.30.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.269398009957513e-06, "max_abs_err": 0.076171875 }, { "name": "model.language_model.layers.30.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.4744570055190707e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.30.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.9825677100016037e-06, "max_abs_err": 0.0177001953125 }, { "name": "model.language_model.layers.30.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.5920844589345506e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.30.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.268531716254074e-06, "max_abs_err": 0.014892578125 }, { "name": "model.language_model.layers.30.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1224999525438761e-06, "max_abs_err": 0.011474609375 }, { "name": "model.language_model.layers.30.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2369683872748283e-06, "max_abs_err": 0.0130615234375 }, { "name": "model.language_model.layers.30.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1966303645749576e-06, "max_abs_err": 0.029296875 }, { "name": "model.language_model.layers.31.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.1083990304759936e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.31.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.267600621053134e-06, "max_abs_err": 0.02734375 }, { "name": "model.language_model.layers.31.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.8369038318487583e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.31.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.287315737703466e-06, "max_abs_err": 0.083984375 }, { "name": "model.language_model.layers.31.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1038995353374048e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.31.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.240384222001012e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.31.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.20342735954182e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.32.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.254045739391586e-06, "max_abs_err": 0.05859375 }, { "name": "model.language_model.layers.32.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.464838644300471e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.32.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.872336838459887e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.32.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.1451859336375492e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.32.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.5212359560100595e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.32.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0921312423306517e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.32.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2382894283291535e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.32.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2022032933600713e-06, "max_abs_err": 0.029296875 }, { "name": "model.language_model.layers.33.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.286329390699393e-06, "max_abs_err": 0.05859375 }, { "name": "model.language_model.layers.33.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.5343044853798347e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.33.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.8076191281579668e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.33.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.1730882079064031e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.33.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.1475950688909506e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.33.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0870433015952585e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.33.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2483037608035374e-06, "max_abs_err": 0.012451171875 }, { "name": "model.language_model.layers.33.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2077531437171274e-06, "max_abs_err": 0.0224609375 }, { "name": "model.language_model.layers.34.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.233311306554242e-06, "max_abs_err": 0.0478515625 }, { "name": "model.language_model.layers.34.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.479704562574625e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.34.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.9201249870093307e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.34.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.1278132205916336e-06, "max_abs_err": 0.011474609375 }, { "name": "model.language_model.layers.34.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.5102467589022126e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.34.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.0783737707242835e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.34.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2633003052542335e-06, "max_abs_err": 0.0107421875 }, { "name": "model.language_model.layers.34.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2155140893810312e-06, "max_abs_err": 0.03125 }, { "name": "model.language_model.layers.35.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.0276127088436624e-06, "max_abs_err": 0.02685546875 }, { "name": "model.language_model.layers.35.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.1725177248299588e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.35.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 3.12417409986665e-06, "max_abs_err": 0.029296875 }, { "name": "model.language_model.layers.35.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.512040964575135e-06, "max_abs_err": 0.05078125 }, { "name": "model.language_model.layers.35.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1263664418947883e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.35.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2510072338045575e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.35.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2349132703093346e-06, "max_abs_err": 0.0263671875 }, { "name": "model.language_model.layers.36.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2981662368692923e-06, "max_abs_err": 0.04296875 }, { "name": "model.language_model.layers.36.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.5219337658199947e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.36.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.1946291326457867e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.36.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.111264623432362e-06, "max_abs_err": 0.01025390625 }, { "name": "model.language_model.layers.36.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.239805780845927e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.36.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2118916856707074e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.36.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1835695659101475e-06, "max_abs_err": 0.013427734375 }, { "name": "model.language_model.layers.36.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.189924205391435e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.37.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2592519144382095e-06, "max_abs_err": 0.05078125 }, { "name": "model.language_model.layers.37.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.417632686047e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.37.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.4002304144232767e-06, "max_abs_err": 0.0234375 }, { "name": "model.language_model.layers.37.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.2903257129437407e-06, "max_abs_err": 0.010009765625 }, { "name": "model.language_model.layers.37.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.4830660549923778e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.37.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2331579455349129e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.37.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1751166084650322e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.37.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1757222182495752e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.38.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2673586954624625e-06, "max_abs_err": 0.0458984375 }, { "name": "model.language_model.layers.38.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.40560825659486e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.38.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.4235271212091902e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.38.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.0649200703483075e-06, "max_abs_err": 0.00830078125 }, { "name": "model.language_model.layers.38.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.1837465737917228e-06, "max_abs_err": 0.013427734375 }, { "name": "model.language_model.layers.38.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3796383200315177e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.38.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.126497636505519e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.38.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1167862794536632e-06, "max_abs_err": 0.052734375 }, { "name": "model.language_model.layers.39.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.0613347260223236e-06, "max_abs_err": 0.0263671875 }, { "name": "model.language_model.layers.39.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.0325528566900175e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.39.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.7667222184390994e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.39.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3898119252407923e-06, "max_abs_err": 0.052734375 }, { "name": "model.language_model.layers.39.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.334637772743008e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.39.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.155729592028365e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.39.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1463904456832097e-06, "max_abs_err": 0.041015625 }, { "name": "model.language_model.layers.40.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2275066839938518e-06, "max_abs_err": 0.072265625 }, { "name": "model.language_model.layers.40.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3229360976984026e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.40.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.2293290840025293e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.40.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.1269736432950594e-06, "max_abs_err": 0.009521484375 }, { "name": "model.language_model.layers.40.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.654160314501496e-06, "max_abs_err": 0.0126953125 }, { "name": "model.language_model.layers.40.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.368590233141731e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.40.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1497323839648743e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.40.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1393324257369386e-06, "max_abs_err": 0.029296875 }, { "name": "model.language_model.layers.41.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.265617922603269e-06, "max_abs_err": 0.068359375 }, { "name": "model.language_model.layers.41.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.319842451470322e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.41.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.2060080482333433e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.41.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.038680125020619e-06, "max_abs_err": 0.01025390625 }, { "name": "model.language_model.layers.41.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.561860813126259e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.41.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2785136505044647e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.41.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.206691081279132e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.41.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1724022215275909e-06, "max_abs_err": 0.0439453125 }, { "name": "model.language_model.layers.42.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2448857635026798e-06, "max_abs_err": 0.099609375 }, { "name": "model.language_model.layers.42.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3134507500799373e-06, "max_abs_err": 0.0224609375 }, { "name": "model.language_model.layers.42.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.1385726540756878e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.42.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.0410058166598901e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.42.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.6826434148242697e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.42.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2140270655436325e-06, "max_abs_err": 0.02587890625 }, { "name": "model.language_model.layers.42.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2454116813387373e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.42.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.1943484423682094e-06, "max_abs_err": 0.02734375 }, { "name": "model.language_model.layers.43.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.0239604054950178e-06, "max_abs_err": 0.02734375 }, { "name": "model.language_model.layers.43.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 1.929708787429263e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.43.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.7688388399838004e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.43.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2910701318323845e-06, "max_abs_err": 0.0341796875 }, { "name": "model.language_model.layers.43.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1980617955487105e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.43.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.277998194382235e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.43.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2299685749894707e-06, "max_abs_err": 0.0225830078125 }, { "name": "model.language_model.layers.44.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.223239789600484e-06, "max_abs_err": 0.056640625 }, { "name": "model.language_model.layers.44.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.391964699199889e-06, "max_abs_err": 0.0224609375 }, { "name": "model.language_model.layers.44.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.0072429833817296e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.44.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.0742689937615069e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.44.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.2013280158716952e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.44.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.179072228296718e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.44.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2814838328267797e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.44.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2340352668616106e-06, "max_abs_err": 0.0224609375 }, { "name": "model.language_model.layers.45.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2801352770329686e-06, "max_abs_err": 0.0380859375 }, { "name": "model.language_model.layers.45.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3841325855755713e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.45.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.010119033002411e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.45.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.185989844998403e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.45.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.601080157342949e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.45.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1739742831196054e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.45.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2792655752491555e-06, "max_abs_err": 0.0166015625 }, { "name": "model.language_model.layers.45.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2291907296457794e-06, "max_abs_err": 0.02880859375 }, { "name": "model.language_model.layers.46.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.204172005804139e-06, "max_abs_err": 0.05859375 }, { "name": "model.language_model.layers.46.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.3992233764147386e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.46.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.0291558939788956e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.46.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.2040126193824108e-06, "max_abs_err": 0.01025390625 }, { "name": "model.language_model.layers.46.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.788545546332898e-06, "max_abs_err": 0.010498046875 }, { "name": "model.language_model.layers.46.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1469211358416942e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.46.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2837932672482566e-06, "max_abs_err": 0.0146484375 }, { "name": "model.language_model.layers.46.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2323175724304747e-06, "max_abs_err": 0.0240478515625 }, { "name": "model.language_model.layers.47.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 1.9556719053070992e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.47.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 2.0843469883402577e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.47.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 3.605088522817823e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.47.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.4847563508956227e-06, "max_abs_err": 0.0322265625 }, { "name": "model.language_model.layers.47.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.159187377197668e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.47.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2961608035766403e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.47.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2382259910737048e-06, "max_abs_err": 0.0247802734375 }, { "name": "model.language_model.layers.48.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.1870594082429307e-06, "max_abs_err": 0.0439453125 }, { "name": "model.language_model.layers.48.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.419677457510261e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.48.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.0048960323038045e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.48.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 8.597081659900141e-07, "max_abs_err": 0.010009765625 }, { "name": "model.language_model.layers.48.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.246070380351739e-06, "max_abs_err": 0.01611328125 }, { "name": "model.language_model.layers.48.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1489769349282142e-06, "max_abs_err": 0.0263671875 }, { "name": "model.language_model.layers.48.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3015267086302629e-06, "max_abs_err": 0.012451171875 }, { "name": "model.language_model.layers.48.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2402294942148728e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.49.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.2726153474650346e-06, "max_abs_err": 0.044921875 }, { "name": "model.language_model.layers.49.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.579964984761318e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.49.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 1.9097521999356104e-06, "max_abs_err": 0.01416015625 }, { "name": "model.language_model.layers.49.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 9.304934565079748e-07, "max_abs_err": 0.00927734375 }, { "name": "model.language_model.layers.49.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.82050962393987e-06, "max_abs_err": 0.01123046875 }, { "name": "model.language_model.layers.49.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1480617558845552e-06, "max_abs_err": 0.02734375 }, { "name": "model.language_model.layers.49.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3112282886140747e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.49.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2458167475415394e-06, "max_abs_err": 0.0322265625 }, { "name": "model.language_model.layers.50.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.1664484393113526e-06, "max_abs_err": 0.05078125 }, { "name": "model.language_model.layers.50.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.5091596853599185e-06, "max_abs_err": 0.0205078125 }, { "name": "model.language_model.layers.50.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.0283023332012817e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.50.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 8.518659342371393e-07, "max_abs_err": 0.0087890625 }, { "name": "model.language_model.layers.50.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.0717093320854474e-06, "max_abs_err": 0.010498046875 }, { "name": "model.language_model.layers.50.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.1787027460741228e-06, "max_abs_err": 0.0263671875 }, { "name": "model.language_model.layers.50.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3201912452132092e-06, "max_abs_err": 0.01220703125 }, { "name": "model.language_model.layers.50.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.240662300006079e-06, "max_abs_err": 0.03515625 }, { "name": "model.language_model.layers.51.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 1.9132123725285055e-06, "max_abs_err": 0.02392578125 }, { "name": "model.language_model.layers.51.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 1.946744760061847e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.51.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 4.022108896606369e-06, "max_abs_err": 0.041015625 }, { "name": "model.language_model.layers.51.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.8496244794951053e-06, "max_abs_err": 0.05859375 }, { "name": "model.language_model.layers.51.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.2004348945993115e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.51.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.356215875603084e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.51.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2943698948220117e-06, "max_abs_err": 0.0264892578125 }, { "name": "model.language_model.layers.52.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.20886636270734e-06, "max_abs_err": 0.033935546875 }, { "name": "model.language_model.layers.52.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.7478515676193638e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.52.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.2069925762480125e-06, "max_abs_err": 0.018798828125 }, { "name": "model.language_model.layers.52.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 9.088012689062452e-07, "max_abs_err": 0.008544921875 }, { "name": "model.language_model.layers.52.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.3718043848930392e-06, "max_abs_err": 0.011962890625 }, { "name": "model.language_model.layers.52.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3292890344018815e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.52.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.313117309109657e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.52.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.277051524084527e-06, "max_abs_err": 0.026123046875 }, { "name": "model.language_model.layers.53.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.268584466946777e-06, "max_abs_err": 0.0419921875 }, { "name": "model.language_model.layers.53.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.724445494095562e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.53.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.277671683259541e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.53.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.107195885197143e-06, "max_abs_err": 0.0087890625 }, { "name": "model.language_model.layers.53.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.592599230410997e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.53.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3979613413539482e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.53.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.300540816373541e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.53.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2685982255788986e-06, "max_abs_err": 0.0235595703125 }, { "name": "model.language_model.layers.54.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.290187467224314e-06, "max_abs_err": 0.0419921875 }, { "name": "model.language_model.layers.54.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.6050677206512773e-06, "max_abs_err": 0.0224609375 }, { "name": "model.language_model.layers.54.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.4078124170046067e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.54.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 9.498465942670009e-07, "max_abs_err": 0.00927734375 }, { "name": "model.language_model.layers.54.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.4075984583760146e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.54.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.5115974747459404e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.54.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3183001783545478e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.54.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2775559525834979e-06, "max_abs_err": 0.046875 }, { "name": "model.language_model.layers.55.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 2.043426547970739e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.55.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 1.8406939261694788e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.55.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 4.5312067413760815e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.55.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.98665167974832e-06, "max_abs_err": 0.087890625 }, { "name": "model.language_model.layers.55.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.4718385727974237e-06, "max_abs_err": 0.01904296875 }, { "name": "model.language_model.layers.55.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3311366728885332e-06, "max_abs_err": 0.0103759765625 }, { "name": "model.language_model.layers.55.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.3066988913124078e-06, "max_abs_err": 0.0286865234375 }, { "name": "model.language_model.layers.56.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.289907342856168e-06, "max_abs_err": 0.0517578125 }, { "name": "model.language_model.layers.56.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.775112989183981e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.56.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3651946321479045e-06, "max_abs_err": 0.01513671875 }, { "name": "model.language_model.layers.56.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.1540782907104585e-06, "max_abs_err": 0.009033203125 }, { "name": "model.language_model.layers.56.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.7983496693195775e-06, "max_abs_err": 0.01318359375 }, { "name": "model.language_model.layers.56.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.5355552704932052e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.56.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3453335441226955e-06, "max_abs_err": 0.011962890625 }, { "name": "model.language_model.layers.56.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.324012146142195e-06, "max_abs_err": 0.0283203125 }, { "name": "model.language_model.layers.57.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3652751224290114e-06, "max_abs_err": 0.04296875 }, { "name": "model.language_model.layers.57.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.8055339953425573e-06, "max_abs_err": 0.02880859375 }, { "name": "model.language_model.layers.57.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.430956101306947e-06, "max_abs_err": 0.017578125 }, { "name": "model.language_model.layers.57.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.1819214478236972e-06, "max_abs_err": 0.009033203125 }, { "name": "model.language_model.layers.57.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.6519981020101113e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.57.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.6104850146803074e-06, "max_abs_err": 0.0206298828125 }, { "name": "model.language_model.layers.57.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.3839515986546758e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.57.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.3615108400699683e-06, "max_abs_err": 0.0274658203125 }, { "name": "model.language_model.layers.58.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.358176743655349e-06, "max_abs_err": 0.043701171875 }, { "name": "model.language_model.layers.58.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.725602826103568e-06, "max_abs_err": 0.02392578125 }, { "name": "model.language_model.layers.58.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.542141601225012e-06, "max_abs_err": 0.01708984375 }, { "name": "model.language_model.layers.58.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.2194138889753958e-06, "max_abs_err": 0.009765625 }, { "name": "model.language_model.layers.58.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.8080114589101868e-06, "max_abs_err": 0.01806640625 }, { "name": "model.language_model.layers.58.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.6701671938790241e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.58.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.4344351484396611e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.58.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.415632482348883e-06, "max_abs_err": 0.0302734375 }, { "name": "model.language_model.layers.59.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 1.9353001334820874e-06, "max_abs_err": 0.02294921875 }, { "name": "model.language_model.layers.59.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 1.6988777815640788e-06, "max_abs_err": 0.01953125 }, { "name": "model.language_model.layers.59.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 6.999664947215933e-06, "max_abs_err": 0.048828125 }, { "name": "model.language_model.layers.59.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 3.5396255952946376e-06, "max_abs_err": 0.0517578125 }, { "name": "model.language_model.layers.59.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.6391746839872212e-06, "max_abs_err": 0.02001953125 }, { "name": "model.language_model.layers.59.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.4502899148283177e-06, "max_abs_err": 0.01263427734375 }, { "name": "model.language_model.layers.59.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.4242756378735066e-06, "max_abs_err": 0.032958984375 }, { "name": "model.language_model.layers.60.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.3661248178541427e-06, "max_abs_err": 0.060546875 }, { "name": "model.language_model.layers.60.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.916745415859623e-06, "max_abs_err": 0.0390625 }, { "name": "model.language_model.layers.60.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3841189431550447e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.60.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 8.732433798286365e-07, "max_abs_err": 0.008544921875 }, { "name": "model.language_model.layers.60.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 2.5511747026030207e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.60.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.620248781364353e-06, "max_abs_err": 0.02490234375 }, { "name": "model.language_model.layers.60.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.4750023638043785e-06, "max_abs_err": 0.01171875 }, { "name": "model.language_model.layers.60.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.4333785429698764e-06, "max_abs_err": 0.044921875 }, { "name": "model.language_model.layers.61.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.4452349407511065e-06, "max_abs_err": 0.048828125 }, { "name": "model.language_model.layers.61.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.5914139314409113e-06, "max_abs_err": 0.0244140625 }, { "name": "model.language_model.layers.61.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.5164081307593733e-06, "max_abs_err": 0.02392578125 }, { "name": "model.language_model.layers.61.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 1.150838784269581e-06, "max_abs_err": 0.014404296875 }, { "name": "model.language_model.layers.61.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.1853935524850385e-06, "max_abs_err": 0.015625 }, { "name": "model.language_model.layers.61.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.6425234434791491e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.61.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.5238819059959496e-06, "max_abs_err": 0.0142822265625 }, { "name": "model.language_model.layers.61.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.4673765917905257e-06, "max_abs_err": 0.0177001953125 }, { "name": "model.language_model.layers.62.linear_attn.out_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.530824076529825e-06, "max_abs_err": 0.046875 }, { "name": "model.language_model.layers.62.linear_attn.in_proj_qkv", "shape": [ 10240, 5120 ], "dtype": "bfloat16", "mse": 2.704540975173586e-06, "max_abs_err": 0.0322265625 }, { "name": "model.language_model.layers.62.linear_attn.in_proj_z", "shape": [ 6144, 5120 ], "dtype": "bfloat16", "mse": 2.3962290924828267e-06, "max_abs_err": 0.021484375 }, { "name": "model.language_model.layers.62.linear_attn.in_proj_b", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 9.156046303360199e-07, "max_abs_err": 0.01025390625 }, { "name": "model.language_model.layers.62.linear_attn.in_proj_a", "shape": [ 48, 5120 ], "dtype": "bfloat16", "mse": 3.0961543870944297e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.62.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.6535709619347472e-06, "max_abs_err": 0.0185546875 }, { "name": "model.language_model.layers.62.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.5455027551070089e-06, "max_abs_err": 0.013671875 }, { "name": "model.language_model.layers.62.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.4387295550477575e-06, "max_abs_err": 0.03125 }, { "name": "model.language_model.layers.63.self_attn.q_proj", "shape": [ 12288, 5120 ], "dtype": "bfloat16", "mse": 1.940002448463929e-06, "max_abs_err": 0.025390625 }, { "name": "model.language_model.layers.63.self_attn.k_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 1.774317979652551e-06, "max_abs_err": 0.02197265625 }, { "name": "model.language_model.layers.63.self_attn.v_proj", "shape": [ 1024, 5120 ], "dtype": "bfloat16", "mse": 6.513901553262258e-06, "max_abs_err": 0.0439453125 }, { "name": "model.language_model.layers.63.self_attn.o_proj", "shape": [ 5120, 6144 ], "dtype": "bfloat16", "mse": 2.9778173029626487e-06, "max_abs_err": 0.0517578125 }, { "name": "model.language_model.layers.63.mlp.gate_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.7987732690016855e-06, "max_abs_err": 0.02099609375 }, { "name": "model.language_model.layers.63.mlp.up_proj", "shape": [ 17408, 5120 ], "dtype": "bfloat16", "mse": 1.6323759837177931e-06, "max_abs_err": 0.02880859375 }, { "name": "model.language_model.layers.63.mlp.down_proj", "shape": [ 5120, 17408 ], "dtype": "bfloat16", "mse": 1.2651813676711754e-06, "max_abs_err": 0.033203125 } ], "source_model": "Qwen/Qwen3.6-27B", "dtype": "bf16", "target_regex": "^model\\.language_model\\..*(?:q_proj|k_proj|v_proj|o_proj|gate_proj|up_proj|down_proj|in_proj_qkv|in_proj_z|in_proj_a|in_proj_b|out_proj)$", "ignore_regex": [], "include_lm_head": false, "model_class": "image-text-to-text", "note": "Dense fake quant: rounded weights are saved as normal HF tensors." }