{ "format": "mxfp4", "codebook": "fp4-e2m1", "block_size": 32, "num_modules": 252, "num_weights": 3633315840, "elapsed_sec": 92.24, "quantized_modules": [ { "name": "model.layers.0.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.264279479888501e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.0.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.839795903739287e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.0.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.153950041858479e-06, "max_abs_err": 0.01953125 }, { "name": "model.layers.0.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.610793894244125e-06, "max_abs_err": 0.068359375 }, { "name": "model.layers.0.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.454409347294131e-06, "max_abs_err": 0.044921875 }, { "name": "model.layers.0.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.75333581562154e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.0.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.2775776566704735e-06, "max_abs_err": 0.0498046875 }, { "name": "model.layers.1.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.5719366426055785e-06, "max_abs_err": 0.05078125 }, { "name": "model.layers.1.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.943770818499615e-06, "max_abs_err": 0.0283203125 }, { "name": "model.layers.1.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.28392683918355e-06, "max_abs_err": 0.01904296875 }, { "name": "model.layers.1.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.713726411864627e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.1.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.142260761203943e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.1.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 2.3373838757834164e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.1.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 2.67392783825926e-06, "max_abs_err": 0.103515625 }, { "name": "model.layers.2.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.887651357421419e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.2.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.243308805802371e-06, "max_abs_err": 0.0498046875 }, { "name": "model.layers.2.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.689682893716963e-06, "max_abs_err": 0.017578125 }, { "name": "model.layers.2.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.92796789330896e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.2.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 3.4752611099975184e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.2.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 2.9105626708769705e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.2.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 2.901462949012057e-06, "max_abs_err": 0.060546875 }, { "name": "model.layers.3.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.9196141844731756e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.3.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.927661848341813e-06, "max_abs_err": 0.02880859375 }, { "name": "model.layers.3.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.769296876678709e-06, "max_abs_err": 0.02197265625 }, { "name": "model.layers.3.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.989236913388595e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.3.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.198741630214499e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.3.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 3.344088327139616e-06, "max_abs_err": 0.0517578125 }, { "name": "model.layers.3.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 3.5842083434545202e-06, "max_abs_err": 0.0625 }, { "name": "model.layers.4.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.097820121591212e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.4.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.2020213843206875e-06, "max_abs_err": 0.0341796875 }, { "name": "model.layers.4.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.889538897463353e-06, "max_abs_err": 0.017578125 }, { "name": "model.layers.4.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.221801984589547e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.4.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 9.631655302655417e-06, "max_abs_err": 0.0380859375 }, { "name": "model.layers.4.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.0591639844933525e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.4.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.182451903034234e-06, "max_abs_err": 0.0595703125 }, { "name": "model.layers.5.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.0453932090022136e-06, "max_abs_err": 0.0380859375 }, { "name": "model.layers.5.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.01126214658143e-06, "max_abs_err": 0.0439453125 }, { "name": "model.layers.5.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.791321200376842e-06, "max_abs_err": 0.01953125 }, { "name": "model.layers.5.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.863534741161857e-06, "max_abs_err": 0.05859375 }, { "name": "model.layers.5.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.399080797971692e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.5.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.56033376394771e-06, "max_abs_err": 0.044921875 }, { "name": "model.layers.5.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.451659151527565e-06, "max_abs_err": 0.083984375 }, { "name": "model.layers.6.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.3390394896268845e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.6.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.041272743255831e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.6.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.7908710004994646e-06, "max_abs_err": 0.0322265625 }, { "name": "model.layers.6.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.193893230170943e-06, "max_abs_err": 0.083984375 }, { "name": "model.layers.6.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.463393558282405e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.6.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.825632458960172e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.6.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.796184839506168e-06, "max_abs_err": 0.0556640625 }, { "name": "model.layers.7.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.0428302529326174e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.7.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.070142833574209e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.7.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.636859325226396e-06, "max_abs_err": 0.021484375 }, { "name": "model.layers.7.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.062678610556759e-06, "max_abs_err": 0.05859375 }, { "name": "model.layers.7.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.295619980141055e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.7.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.886428996542236e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.7.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.855854513152735e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.8.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.100348971609492e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.8.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.225751465331996e-06, "max_abs_err": 0.0263671875 }, { "name": "model.layers.8.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.604257123399293e-06, "max_abs_err": 0.01806640625 }, { "name": "model.layers.8.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.294675702316454e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.8.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.334666068141814e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.8.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.1152342166460585e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.8.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.144322130945511e-06, "max_abs_err": 0.125 }, { "name": "model.layers.9.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.864109087066026e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.9.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.887382601737045e-06, "max_abs_err": 0.0439453125 }, { "name": "model.layers.9.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.763327291992027e-06, "max_abs_err": 0.02099609375 }, { "name": "model.layers.9.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.205638899496989e-06, "max_abs_err": 0.0478515625 }, { "name": "model.layers.9.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.41132771509001e-06, "max_abs_err": 0.056640625 }, { "name": "model.layers.9.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.700968474935507e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.9.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.611386884789681e-06, "max_abs_err": 0.060546875 }, { "name": "model.layers.10.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.169024007045664e-06, "max_abs_err": 0.0361328125 }, { "name": "model.layers.10.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.15500187248108e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.10.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.621941338380566e-06, "max_abs_err": 0.0234375 }, { "name": "model.layers.10.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.404429430200253e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.10.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.087458925525425e-06, "max_abs_err": 0.07421875 }, { "name": "model.layers.10.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.886557235295186e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.10.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.850164714298444e-06, "max_abs_err": 0.080078125 }, { "name": "model.layers.11.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.154139671503799e-06, "max_abs_err": 0.0380859375 }, { "name": "model.layers.11.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.333734861778794e-06, "max_abs_err": 0.0517578125 }, { "name": "model.layers.11.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.783987828384852e-06, "max_abs_err": 0.021484375 }, { "name": "model.layers.11.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.283522568788612e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.11.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.638977083639475e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.11.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.182515906199114e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.11.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.072423391538905e-06, "max_abs_err": 0.0458984375 }, { "name": "model.layers.12.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.248842626315309e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.12.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.132031674293103e-06, "max_abs_err": 0.0625 }, { "name": "model.layers.12.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.6453258114052005e-06, "max_abs_err": 0.02490234375 }, { "name": "model.layers.12.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.179861091164639e-06, "max_abs_err": 0.091796875 }, { "name": "model.layers.12.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.296268111327663e-06, "max_abs_err": 0.0615234375 }, { "name": "model.layers.12.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.405216597864637e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.12.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.189312105358113e-06, "max_abs_err": 0.0576171875 }, { "name": "model.layers.13.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.209337359701749e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.13.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.068667633167934e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.13.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.373182259267196e-06, "max_abs_err": 0.021484375 }, { "name": "model.layers.13.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.950931270286674e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.13.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.874879661860177e-06, "max_abs_err": 0.076171875 }, { "name": "model.layers.13.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.544491614273284e-06, "max_abs_err": 0.0380859375 }, { "name": "model.layers.13.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.24446249983157e-06, "max_abs_err": 0.08203125 }, { "name": "model.layers.14.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.205360594118247e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.14.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.147089723323006e-06, "max_abs_err": 0.0322265625 }, { "name": "model.layers.14.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.03566240897635e-06, "max_abs_err": 0.01953125 }, { "name": "model.layers.14.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.813705345441122e-06, "max_abs_err": 0.076171875 }, { "name": "model.layers.14.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.712263828172581e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.14.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.389957550505642e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.14.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.130104000272695e-06, "max_abs_err": 0.09375 }, { "name": "model.layers.15.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.329152827471262e-06, "max_abs_err": 0.0439453125 }, { "name": "model.layers.15.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.094994776300155e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.15.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.915267593081808e-06, "max_abs_err": 0.0283203125 }, { "name": "model.layers.15.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.53751044915407e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.15.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.323084678821033e-06, "max_abs_err": 0.068359375 }, { "name": "model.layers.15.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.271551799523877e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.15.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.047445029049413e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.16.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.020708158554044e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.16.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.788883870787686e-06, "max_abs_err": 0.0322265625 }, { "name": "model.layers.16.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.932683961873408e-06, "max_abs_err": 0.0205078125 }, { "name": "model.layers.16.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.804500804311829e-06, "max_abs_err": 0.078125 }, { "name": "model.layers.16.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.588532076217234e-06, "max_abs_err": 0.078125 }, { "name": "model.layers.16.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.20666526426794e-06, "max_abs_err": 0.12109375 }, { "name": "model.layers.16.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.8864503696677275e-06, "max_abs_err": 0.1484375 }, { "name": "model.layers.17.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.277245691104326e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.17.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.526109478319995e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.17.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.642662927130004e-06, "max_abs_err": 0.0244140625 }, { "name": "model.layers.17.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.645903572964016e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.17.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.43499709237949e-06, "max_abs_err": 0.1015625 }, { "name": "model.layers.17.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.242377483227756e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.17.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.918028935207985e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.18.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.925951543555129e-06, "max_abs_err": 0.0537109375 }, { "name": "model.layers.18.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.623555014404701e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.18.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.86613441555528e-06, "max_abs_err": 0.02197265625 }, { "name": "model.layers.18.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.651018116419436e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.18.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.3064359235577285e-06, "max_abs_err": 0.0859375 }, { "name": "model.layers.18.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.234990112512605e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.18.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.910174538963474e-06, "max_abs_err": 0.072265625 }, { "name": "model.layers.19.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.110360689286608e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.19.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.34439562013722e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.19.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.596968094323529e-06, "max_abs_err": 0.0224609375 }, { "name": "model.layers.19.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.723720394395059e-06, "max_abs_err": 0.08203125 }, { "name": "model.layers.19.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.240663085714914e-06, "max_abs_err": 0.08203125 }, { "name": "model.layers.19.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.265425897960085e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.19.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.908146820525872e-06, "max_abs_err": 0.07421875 }, { "name": "model.layers.20.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.4594352150161285e-06, "max_abs_err": 0.0458984375 }, { "name": "model.layers.20.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.236061613482889e-06, "max_abs_err": 0.0341796875 }, { "name": "model.layers.20.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.584740392805543e-06, "max_abs_err": 0.02294921875 }, { "name": "model.layers.20.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.505439392232802e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.20.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.135244009579765e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.20.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.288481133902678e-06, "max_abs_err": 0.05078125 }, { "name": "model.layers.20.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 4.983662165614078e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.21.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.224626875133254e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.21.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.212050498608733e-06, "max_abs_err": 0.029296875 }, { "name": "model.layers.21.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.426644293038407e-06, "max_abs_err": 0.0361328125 }, { "name": "model.layers.21.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.37218432125519e-06, "max_abs_err": 0.0302734375 }, { "name": "model.layers.21.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.967777385900263e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.21.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.256719305180013e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.21.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.050899289926747e-06, "max_abs_err": 0.080078125 }, { "name": "model.layers.22.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.0893240768346e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.22.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 3.946858214476379e-06, "max_abs_err": 0.10546875 }, { "name": "model.layers.22.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.398288976721233e-06, "max_abs_err": 0.02734375 }, { "name": "model.layers.22.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.562843969324604e-06, "max_abs_err": 0.0576171875 }, { "name": "model.layers.22.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.315303496900015e-06, "max_abs_err": 0.056640625 }, { "name": "model.layers.22.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.501461146195652e-06, "max_abs_err": 0.06640625 }, { "name": "model.layers.22.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.105726813781075e-06, "max_abs_err": 0.05859375 }, { "name": "model.layers.23.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.768430128227919e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.23.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.367841029306874e-06, "max_abs_err": 0.0380859375 }, { "name": "model.layers.23.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.622457138088066e-06, "max_abs_err": 0.025390625 }, { "name": "model.layers.23.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.904564775642939e-06, "max_abs_err": 0.0478515625 }, { "name": "model.layers.23.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.713046448363457e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.23.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.536606295208912e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.23.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.168157713342225e-06, "max_abs_err": 0.0859375 }, { "name": "model.layers.24.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.856135430803988e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.24.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.3650697989505716e-06, "max_abs_err": 0.0283203125 }, { "name": "model.layers.24.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.5182254072860815e-06, "max_abs_err": 0.0361328125 }, { "name": "model.layers.24.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.061527190264314e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.24.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.936073193879565e-06, "max_abs_err": 0.0439453125 }, { "name": "model.layers.24.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.574751867243322e-06, "max_abs_err": 0.0380859375 }, { "name": "model.layers.24.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.253869858279359e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.25.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.219796548772138e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.25.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.655545581044862e-06, "max_abs_err": 0.0283203125 }, { "name": "model.layers.25.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.9699368648580275e-06, "max_abs_err": 0.02587890625 }, { "name": "model.layers.25.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.926278506900417e-06, "max_abs_err": 0.029296875 }, { "name": "model.layers.25.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.071639745641733e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.25.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.578469881584169e-06, "max_abs_err": 0.05078125 }, { "name": "model.layers.25.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.302589670463931e-06, "max_abs_err": 0.068359375 }, { "name": "model.layers.26.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.408423021435738e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.26.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.52501944891992e-06, "max_abs_err": 0.025390625 }, { "name": "model.layers.26.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.899823008803651e-06, "max_abs_err": 0.02490234375 }, { "name": "model.layers.26.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.148473974259105e-06, "max_abs_err": 0.0322265625 }, { "name": "model.layers.26.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.090650913392892e-06, "max_abs_err": 0.0498046875 }, { "name": "model.layers.26.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.705280273105018e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.26.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.437568688648753e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.27.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.309313564794138e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.27.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.194436769466847e-06, "max_abs_err": 0.01953125 }, { "name": "model.layers.27.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.552273250941653e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.27.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.919363618682837e-06, "max_abs_err": 0.0458984375 }, { "name": "model.layers.27.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.0298166317807045e-06, "max_abs_err": 0.060546875 }, { "name": "model.layers.27.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.844984571012901e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.27.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.563001195696415e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.28.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.536449862120207e-06, "max_abs_err": 0.0361328125 }, { "name": "model.layers.28.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.469477062229998e-06, "max_abs_err": 0.02880859375 }, { "name": "model.layers.28.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.15230601397343e-06, "max_abs_err": 0.021484375 }, { "name": "model.layers.28.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 4.8867068471736275e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.28.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.904380486754235e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.28.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.968320692772977e-06, "max_abs_err": 0.05859375 }, { "name": "model.layers.28.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.702905582438689e-06, "max_abs_err": 0.05078125 }, { "name": "model.layers.29.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 6.084032975195441e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.29.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.273008926247712e-06, "max_abs_err": 0.0322265625 }, { "name": "model.layers.29.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.62526020075893e-06, "max_abs_err": 0.0302734375 }, { "name": "model.layers.29.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.26668509337469e-06, "max_abs_err": 0.02734375 }, { "name": "model.layers.29.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.9085555221827235e-06, "max_abs_err": 0.076171875 }, { "name": "model.layers.29.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.158456471894169e-06, "max_abs_err": 0.056640625 }, { "name": "model.layers.29.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.892638000659645e-06, "max_abs_err": 0.056640625 }, { "name": "model.layers.30.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.1339798119443e-06, "max_abs_err": 0.0458984375 }, { "name": "model.layers.30.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.627934231393738e-06, "max_abs_err": 0.03076171875 }, { "name": "model.layers.30.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.30783461019746e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.30.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.175373644306092e-06, "max_abs_err": 0.029296875 }, { "name": "model.layers.30.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.838384822709486e-06, "max_abs_err": 0.087890625 }, { "name": "model.layers.30.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.302471774688456e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.30.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.0402985582186375e-06, "max_abs_err": 0.064453125 }, { "name": "model.layers.31.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.342665190255502e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.31.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.267042641004082e-06, "max_abs_err": 0.0283203125 }, { "name": "model.layers.31.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.811584171373397e-06, "max_abs_err": 0.02685546875 }, { "name": "model.layers.31.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.487268936121836e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.31.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.7160182223015e-06, "max_abs_err": 0.080078125 }, { "name": "model.layers.31.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.454072263295529e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.31.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.2860849538992625e-06, "max_abs_err": 0.060546875 }, { "name": "model.layers.32.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.16586715093581e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.32.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.320722382544773e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.32.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.77997013553977e-06, "max_abs_err": 0.02783203125 }, { "name": "model.layers.32.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.459540261654183e-06, "max_abs_err": 0.0537109375 }, { "name": "model.layers.32.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.551322374230949e-06, "max_abs_err": 0.0546875 }, { "name": "model.layers.32.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.4898877099039964e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.32.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.295555976976175e-06, "max_abs_err": 0.072265625 }, { "name": "model.layers.33.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 5.228690952208126e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.33.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 3.49267543242604e-06, "max_abs_err": 0.025390625 }, { "name": "model.layers.33.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.739745342405513e-06, "max_abs_err": 0.02734375 }, { "name": "model.layers.33.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.923426670051413e-06, "max_abs_err": 0.099609375 }, { "name": "model.layers.33.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.439240794657962e-06, "max_abs_err": 0.068359375 }, { "name": "model.layers.33.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.534968179039424e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.33.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.153291906230152e-06, "max_abs_err": 0.07421875 }, { "name": "model.layers.34.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.863522917730734e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.34.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.038306087750243e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.34.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.699466550548095e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.34.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.641515599563718e-06, "max_abs_err": 0.0498046875 }, { "name": "model.layers.34.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.64941683478537e-06, "max_abs_err": 0.076171875 }, { "name": "model.layers.34.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.364585715346038e-06, "max_abs_err": 0.060546875 }, { "name": "model.layers.34.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.501821306097554e-06, "max_abs_err": 0.08984375 }, { "name": "model.layers.35.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 4.985550276614958e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.35.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 4.587274816003628e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.35.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.806665624026209e-06, "max_abs_err": 0.0263671875 }, { "name": "model.layers.35.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 5.292796686262591e-06, "max_abs_err": 0.078125 }, { "name": "model.layers.35.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.248279078135965e-06, "max_abs_err": 0.08984375 }, { "name": "model.layers.35.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.548268174810801e-06, "max_abs_err": 0.08203125 }, { "name": "model.layers.35.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.007366326026386e-06, "max_abs_err": 0.076171875 } ], "source_model": "Qwen/Qwen3-4B-Thinking-2507", "dtype": "bf16", "target_regex": ".*", "ignore_regex": [], "include_lm_head": false, "note": "Dense fake quant: rounded weights are saved as normal HF tensors." }