{ "format": "fp4-e1m2", "codebook": "fp4-e1m2", "block_size": 128, "num_modules": 252, "num_weights": 3633315840, "elapsed_sec": 84.505, "quantized_modules": [ { "name": "model.layers.0.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.588077096443158e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.0.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.22014862933429e-06, "max_abs_err": 0.0205078125 }, { "name": "model.layers.0.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.377656402240973e-06, "max_abs_err": 0.01177978515625 }, { "name": "model.layers.0.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.650277555309003e-06, "max_abs_err": 0.0361328125 }, { "name": "model.layers.0.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.575779363833135e-06, "max_abs_err": 0.036865234375 }, { "name": "model.layers.0.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.578104148502462e-06, "max_abs_err": 0.0234375 }, { "name": "model.layers.0.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.829102287360001e-06, "max_abs_err": 0.03955078125 }, { "name": "model.layers.1.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 6.567387572431471e-06, "max_abs_err": 0.0267333984375 }, { "name": "model.layers.1.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.986399966990575e-06, "max_abs_err": 0.016357421875 }, { "name": "model.layers.1.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.449831173289567e-06, "max_abs_err": 0.0107421875 }, { "name": "model.layers.1.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.55725034448551e-06, "max_abs_err": 0.02880859375 }, { "name": "model.layers.1.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.097381512721768e-06, "max_abs_err": 0.02734375 }, { "name": "model.layers.1.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 3.2924058359640185e-06, "max_abs_err": 0.020751953125 }, { "name": "model.layers.1.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 3.914849457942182e-06, "max_abs_err": 0.050537109375 }, { "name": "model.layers.2.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.008223292359617e-06, "max_abs_err": 0.0233154296875 }, { "name": "model.layers.2.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.825708962627687e-06, "max_abs_err": 0.0289306640625 }, { "name": "model.layers.2.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.030953722482082e-06, "max_abs_err": 0.01507568359375 }, { "name": "model.layers.2.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.778335773560684e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.2.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.990461548004532e-06, "max_abs_err": 0.021484375 }, { "name": "model.layers.2.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.2477140596020035e-06, "max_abs_err": 0.035888671875 }, { "name": "model.layers.2.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.821545073558809e-06, "max_abs_err": 0.052734375 }, { "name": "model.layers.3.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.1592644417250995e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.3.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.141894457163289e-06, "max_abs_err": 0.0245361328125 }, { "name": "model.layers.3.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.189883374143392e-06, "max_abs_err": 0.013427734375 }, { "name": "model.layers.3.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.849753845017403e-06, "max_abs_err": 0.029052734375 }, { "name": "model.layers.3.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.640875421406236e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.3.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 4.7608114073227625e-06, "max_abs_err": 0.03173828125 }, { "name": "model.layers.3.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 5.866871561011067e-06, "max_abs_err": 0.05126953125 }, { "name": "model.layers.4.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.368921160377795e-06, "max_abs_err": 0.030029296875 }, { "name": "model.layers.4.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.846755579521414e-06, "max_abs_err": 0.052001953125 }, { "name": "model.layers.4.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.242974217864685e-06, "max_abs_err": 0.01318359375 }, { "name": "model.layers.4.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.102682957338402e-06, "max_abs_err": 0.0341796875 }, { "name": "model.layers.4.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 1.3452598068397492e-05, "max_abs_err": 0.0223388671875 }, { "name": "model.layers.4.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 5.699404482584214e-06, "max_abs_err": 0.02587890625 }, { "name": "model.layers.4.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.788152859371621e-06, "max_abs_err": 0.0458984375 }, { "name": "model.layers.5.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.280615136551205e-06, "max_abs_err": 0.031494140625 }, { "name": "model.layers.5.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.214591278170701e-06, "max_abs_err": 0.0252685546875 }, { "name": "model.layers.5.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.211250133172143e-06, "max_abs_err": 0.01171875 }, { "name": "model.layers.5.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.684120307909325e-06, "max_abs_err": 0.02978515625 }, { "name": "model.layers.5.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 1.2245306606928352e-05, "max_abs_err": 0.036865234375 }, { "name": "model.layers.5.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.395664968295023e-06, "max_abs_err": 0.043701171875 }, { "name": "model.layers.5.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.789731742173899e-06, "max_abs_err": 0.04833984375 }, { "name": "model.layers.6.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.780379746691324e-06, "max_abs_err": 0.068359375 }, { "name": "model.layers.6.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.2174093475041445e-06, "max_abs_err": 0.0277099609375 }, { "name": "model.layers.6.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.504290235578083e-06, "max_abs_err": 0.0181884765625 }, { "name": "model.layers.6.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.136653493944323e-06, "max_abs_err": 0.037353515625 }, { "name": "model.layers.6.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 1.052189963957062e-05, "max_abs_err": 0.033203125 }, { "name": "model.layers.6.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.755277354386635e-06, "max_abs_err": 0.03466796875 }, { "name": "model.layers.6.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.221412943181349e-06, "max_abs_err": 0.06982421875 }, { "name": "model.layers.7.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.286022992047947e-06, "max_abs_err": 0.02685546875 }, { "name": "model.layers.7.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.398552497761557e-06, "max_abs_err": 0.034423828125 }, { "name": "model.layers.7.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.32370312511921e-06, "max_abs_err": 0.010986328125 }, { "name": "model.layers.7.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.004656708886614e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.7.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 1.0268748155795038e-05, "max_abs_err": 0.038818359375 }, { "name": "model.layers.7.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.796933121222537e-06, "max_abs_err": 0.03857421875 }, { "name": "model.layers.7.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.031244422250893e-06, "max_abs_err": 0.0673828125 }, { "name": "model.layers.8.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.36622541808174e-06, "max_abs_err": 0.02490234375 }, { "name": "model.layers.8.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.366265435848618e-06, "max_abs_err": 0.02685546875 }, { "name": "model.layers.8.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.208777555613779e-06, "max_abs_err": 0.01025390625 }, { "name": "model.layers.8.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.263519364641979e-06, "max_abs_err": 0.03271484375 }, { "name": "model.layers.8.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.75733167049475e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.8.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.125912361516384e-06, "max_abs_err": 0.037353515625 }, { "name": "model.layers.8.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.599578111694427e-06, "max_abs_err": 0.06640625 }, { "name": "model.layers.9.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.24924302630825e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.9.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.58801161282463e-06, "max_abs_err": 0.05419921875 }, { "name": "model.layers.9.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.748740922077559e-06, "max_abs_err": 0.0111083984375 }, { "name": "model.layers.9.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.212303898995742e-06, "max_abs_err": 0.054443359375 }, { "name": "model.layers.9.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 1.0260140697937459e-05, "max_abs_err": 0.032958984375 }, { "name": "model.layers.9.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.576727173523977e-06, "max_abs_err": 0.031494140625 }, { "name": "model.layers.9.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 6.8758558882109355e-06, "max_abs_err": 0.046630859375 }, { "name": "model.layers.10.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.446252766385442e-06, "max_abs_err": 0.0225830078125 }, { "name": "model.layers.10.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.4731769927893765e-06, "max_abs_err": 0.03466796875 }, { "name": "model.layers.10.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.639672498451546e-06, "max_abs_err": 0.01336669921875 }, { "name": "model.layers.10.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.456033017660957e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.10.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 9.759428394318093e-06, "max_abs_err": 0.048095703125 }, { "name": "model.layers.10.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 6.833535280748038e-06, "max_abs_err": 0.05810546875 }, { "name": "model.layers.10.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.248203473864123e-06, "max_abs_err": 0.07080078125 }, { "name": "model.layers.11.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.557281605841126e-06, "max_abs_err": 0.025390625 }, { "name": "model.layers.11.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.718850611126982e-06, "max_abs_err": 0.025390625 }, { "name": "model.layers.11.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.788002898858394e-06, "max_abs_err": 0.012939453125 }, { "name": "model.layers.11.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.280049430846702e-06, "max_abs_err": 0.052001953125 }, { "name": "model.layers.11.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 9.182514986605383e-06, "max_abs_err": 0.03955078125 }, { "name": "model.layers.11.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.25067457096884e-06, "max_abs_err": 0.042724609375 }, { "name": "model.layers.11.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.483451554435305e-06, "max_abs_err": 0.0615234375 }, { "name": "model.layers.12.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.649958206457086e-06, "max_abs_err": 0.022216796875 }, { "name": "model.layers.12.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.527015895902878e-06, "max_abs_err": 0.0341796875 }, { "name": "model.layers.12.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.648221748648211e-06, "max_abs_err": 0.01513671875 }, { "name": "model.layers.12.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.216715857794043e-06, "max_abs_err": 0.052978515625 }, { "name": "model.layers.12.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.810835424810648e-06, "max_abs_err": 0.047119140625 }, { "name": "model.layers.12.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.65768072596984e-06, "max_abs_err": 0.044189453125 }, { "name": "model.layers.12.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.747158633719664e-06, "max_abs_err": 0.06640625 }, { "name": "model.layers.13.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.684538104513194e-06, "max_abs_err": 0.0255126953125 }, { "name": "model.layers.13.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.671169441891834e-06, "max_abs_err": 0.045654296875 }, { "name": "model.layers.13.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.059229341801256e-06, "max_abs_err": 0.015380859375 }, { "name": "model.layers.13.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.882342859171331e-06, "max_abs_err": 0.02978515625 }, { "name": "model.layers.13.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.31723718874855e-06, "max_abs_err": 0.032958984375 }, { "name": "model.layers.13.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.938918315630872e-06, "max_abs_err": 0.05517578125 }, { "name": "model.layers.13.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.148494998749811e-06, "max_abs_err": 0.06640625 }, { "name": "model.layers.14.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.49942319089314e-06, "max_abs_err": 0.0225830078125 }, { "name": "model.layers.14.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.377564998023445e-06, "max_abs_err": 0.0260009765625 }, { "name": "model.layers.14.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.827355486573651e-06, "max_abs_err": 0.01220703125 }, { "name": "model.layers.14.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.691105227218941e-06, "max_abs_err": 0.035400390625 }, { "name": "model.layers.14.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.066808732110076e-06, "max_abs_err": 0.0400390625 }, { "name": "model.layers.14.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.732277481409255e-06, "max_abs_err": 0.031494140625 }, { "name": "model.layers.14.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.969802936713677e-06, "max_abs_err": 0.07421875 }, { "name": "model.layers.15.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.732205631327815e-06, "max_abs_err": 0.0238037109375 }, { "name": "model.layers.15.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.503636879846454e-06, "max_abs_err": 0.024658203125 }, { "name": "model.layers.15.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.502347554895096e-06, "max_abs_err": 0.01300048828125 }, { "name": "model.layers.15.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.3378856793860905e-06, "max_abs_err": 0.0263671875 }, { "name": "model.layers.15.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.587317668367177e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.15.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.58361511543626e-06, "max_abs_err": 0.0322265625 }, { "name": "model.layers.15.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.028428055695258e-06, "max_abs_err": 0.05810546875 }, { "name": "model.layers.16.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.252264367707539e-06, "max_abs_err": 0.0234375 }, { "name": "model.layers.16.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.902268978592474e-06, "max_abs_err": 0.0301513671875 }, { "name": "model.layers.16.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.575649189879186e-06, "max_abs_err": 0.01092529296875 }, { "name": "model.layers.16.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.674705218756571e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.16.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.916810318420175e-06, "max_abs_err": 0.038330078125 }, { "name": "model.layers.16.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.4859863161691464e-06, "max_abs_err": 0.0703125 }, { "name": "model.layers.16.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.672114406886976e-06, "max_abs_err": 0.0859375 }, { "name": "model.layers.17.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.636505870323163e-06, "max_abs_err": 0.023193359375 }, { "name": "model.layers.17.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.5802596509456635e-06, "max_abs_err": 0.0185546875 }, { "name": "model.layers.17.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.191649612854235e-06, "max_abs_err": 0.012939453125 }, { "name": "model.layers.17.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.4882333390414715e-06, "max_abs_err": 0.03466796875 }, { "name": "model.layers.17.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.701862159592565e-06, "max_abs_err": 0.045654296875 }, { "name": "model.layers.17.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.466232545993989e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.17.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.766473572701216e-06, "max_abs_err": 0.050048828125 }, { "name": "model.layers.18.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.10130689185462e-06, "max_abs_err": 0.02392578125 }, { "name": "model.layers.18.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.6081765908165835e-06, "max_abs_err": 0.018310546875 }, { "name": "model.layers.18.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.363783308595885e-06, "max_abs_err": 0.012451171875 }, { "name": "model.layers.18.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.456766186602181e-06, "max_abs_err": 0.02880859375 }, { "name": "model.layers.18.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.610601642227266e-06, "max_abs_err": 0.04541015625 }, { "name": "model.layers.18.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.534623364335857e-06, "max_abs_err": 0.04052734375 }, { "name": "model.layers.18.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.861345693527255e-06, "max_abs_err": 0.049072265625 }, { "name": "model.layers.19.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.459581865987275e-06, "max_abs_err": 0.02734375 }, { "name": "model.layers.19.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.341940206766594e-06, "max_abs_err": 0.02880859375 }, { "name": "model.layers.19.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.081698979367502e-06, "max_abs_err": 0.01202392578125 }, { "name": "model.layers.19.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.664553893642733e-06, "max_abs_err": 0.039306640625 }, { "name": "model.layers.19.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.546796496171737e-06, "max_abs_err": 0.045654296875 }, { "name": "model.layers.19.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.59606064093532e-06, "max_abs_err": 0.06201171875 }, { "name": "model.layers.19.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.922761142253876e-06, "max_abs_err": 0.046630859375 }, { "name": "model.layers.20.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.91044112702366e-06, "max_abs_err": 0.024169921875 }, { "name": "model.layers.20.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.348703209368978e-06, "max_abs_err": 0.02294921875 }, { "name": "model.layers.20.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.16026749816956e-06, "max_abs_err": 0.012939453125 }, { "name": "model.layers.20.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.272050995903555e-06, "max_abs_err": 0.0296630859375 }, { "name": "model.layers.20.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.312453362828819e-06, "max_abs_err": 0.03515625 }, { "name": "model.layers.20.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.526746230723802e-06, "max_abs_err": 0.048583984375 }, { "name": "model.layers.20.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.866769010433927e-06, "max_abs_err": 0.03173828125 }, { "name": "model.layers.21.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.602071036671987e-06, "max_abs_err": 0.0247802734375 }, { "name": "model.layers.21.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.172195753606502e-06, "max_abs_err": 0.0169677734375 }, { "name": "model.layers.21.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 7.992393875611015e-06, "max_abs_err": 0.0196533203125 }, { "name": "model.layers.21.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.150513854663586e-06, "max_abs_err": 0.038330078125 }, { "name": "model.layers.21.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.164737326093018e-06, "max_abs_err": 0.047119140625 }, { "name": "model.layers.21.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.463430392817827e-06, "max_abs_err": 0.0284423828125 }, { "name": "model.layers.21.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.913985427876469e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.22.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.305733106477419e-06, "max_abs_err": 0.0225830078125 }, { "name": "model.layers.22.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.093285719543928e-06, "max_abs_err": 0.05078125 }, { "name": "model.layers.22.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.182991223293357e-06, "max_abs_err": 0.0177001953125 }, { "name": "model.layers.22.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.399286121450132e-06, "max_abs_err": 0.033203125 }, { "name": "model.layers.22.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.68175050325226e-06, "max_abs_err": 0.051513671875 }, { "name": "model.layers.22.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.830390131857712e-06, "max_abs_err": 0.06494140625 }, { "name": "model.layers.22.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.186118066078052e-06, "max_abs_err": 0.060546875 }, { "name": "model.layers.23.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 8.401989362027962e-06, "max_abs_err": 0.0291748046875 }, { "name": "model.layers.23.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.425516858143965e-06, "max_abs_err": 0.0185546875 }, { "name": "model.layers.23.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.733589311304968e-06, "max_abs_err": 0.01458740234375 }, { "name": "model.layers.23.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.869345270388294e-06, "max_abs_err": 0.0244140625 }, { "name": "model.layers.23.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.15053590486059e-06, "max_abs_err": 0.034912109375 }, { "name": "model.layers.23.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.769728654238861e-06, "max_abs_err": 0.0250244140625 }, { "name": "model.layers.23.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.009847988432739e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.24.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 8.559223715565167e-06, "max_abs_err": 0.0245361328125 }, { "name": "model.layers.24.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.481920536316466e-06, "max_abs_err": 0.0172119140625 }, { "name": "model.layers.24.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.675982826389372e-06, "max_abs_err": 0.0181884765625 }, { "name": "model.layers.24.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.029773314570775e-06, "max_abs_err": 0.0244140625 }, { "name": "model.layers.24.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.404917025472969e-06, "max_abs_err": 0.034423828125 }, { "name": "model.layers.24.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.771947821311187e-06, "max_abs_err": 0.026123046875 }, { "name": "model.layers.24.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.879266377130989e-06, "max_abs_err": 0.043701171875 }, { "name": "model.layers.25.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.41469284548657e-06, "max_abs_err": 0.021240234375 }, { "name": "model.layers.25.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.615851361857494e-06, "max_abs_err": 0.017578125 }, { "name": "model.layers.25.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.632548087916803e-06, "max_abs_err": 0.01397705078125 }, { "name": "model.layers.25.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.726159426762024e-06, "max_abs_err": 0.0133056640625 }, { "name": "model.layers.25.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.552839062758721e-06, "max_abs_err": 0.058349609375 }, { "name": "model.layers.25.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.761056622257456e-06, "max_abs_err": 0.02734375 }, { "name": "model.layers.25.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.87825592851732e-06, "max_abs_err": 0.03271484375 }, { "name": "model.layers.26.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.792742508172523e-06, "max_abs_err": 0.0224609375 }, { "name": "model.layers.26.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.506360477942508e-06, "max_abs_err": 0.0174560546875 }, { "name": "model.layers.26.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.668377631693147e-06, "max_abs_err": 0.01287841796875 }, { "name": "model.layers.26.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.0870205490791705e-06, "max_abs_err": 0.016357421875 }, { "name": "model.layers.26.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.496267582813744e-06, "max_abs_err": 0.0419921875 }, { "name": "model.layers.26.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.873268259572797e-06, "max_abs_err": 0.0556640625 }, { "name": "model.layers.26.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.81321432441473e-06, "max_abs_err": 0.04931640625 }, { "name": "model.layers.27.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.498706509068143e-06, "max_abs_err": 0.0269775390625 }, { "name": "model.layers.27.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.990749286866048e-06, "max_abs_err": 0.0167236328125 }, { "name": "model.layers.27.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.322511348524131e-06, "max_abs_err": 0.015869140625 }, { "name": "model.layers.27.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.8053182076255325e-06, "max_abs_err": 0.0205078125 }, { "name": "model.layers.27.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.416613127337769e-06, "max_abs_err": 0.034423828125 }, { "name": "model.layers.27.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.054826139414217e-06, "max_abs_err": 0.0308837890625 }, { "name": "model.layers.27.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 7.993357030500192e-06, "max_abs_err": 0.034912109375 }, { "name": "model.layers.28.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.742137313471176e-06, "max_abs_err": 0.0240478515625 }, { "name": "model.layers.28.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.365925855789101e-06, "max_abs_err": 0.01611328125 }, { "name": "model.layers.28.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.888496267900337e-06, "max_abs_err": 0.011962890625 }, { "name": "model.layers.28.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 6.692939678032417e-06, "max_abs_err": 0.02490234375 }, { "name": "model.layers.28.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.24799371912377e-06, "max_abs_err": 0.041015625 }, { "name": "model.layers.28.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.215581146941986e-06, "max_abs_err": 0.046630859375 }, { "name": "model.layers.28.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.148559572873637e-06, "max_abs_err": 0.05224609375 }, { "name": "model.layers.29.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 8.696166332811117e-06, "max_abs_err": 0.0255126953125 }, { "name": "model.layers.29.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.333238616207382e-06, "max_abs_err": 0.02099609375 }, { "name": "model.layers.29.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 8.486220394843258e-06, "max_abs_err": 0.0230712890625 }, { "name": "model.layers.29.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.236486453621183e-06, "max_abs_err": 0.015625 }, { "name": "model.layers.29.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.153269845934119e-06, "max_abs_err": 0.048583984375 }, { "name": "model.layers.29.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.432884897047188e-06, "max_abs_err": 0.027587890625 }, { "name": "model.layers.29.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.19538126961561e-06, "max_abs_err": 0.03759765625 }, { "name": "model.layers.30.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.331714641622966e-06, "max_abs_err": 0.034423828125 }, { "name": "model.layers.30.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.664628926955629e-06, "max_abs_err": 0.02099609375 }, { "name": "model.layers.30.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.350965228804853e-06, "max_abs_err": 0.017822265625 }, { "name": "model.layers.30.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.0970017986837775e-06, "max_abs_err": 0.017578125 }, { "name": "model.layers.30.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.069856448855717e-06, "max_abs_err": 0.041748046875 }, { "name": "model.layers.30.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.623426765552722e-06, "max_abs_err": 0.027587890625 }, { "name": "model.layers.30.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.439507837465499e-06, "max_abs_err": 0.050537109375 }, { "name": "model.layers.31.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.627363629580941e-06, "max_abs_err": 0.028076171875 }, { "name": "model.layers.31.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.258929261093726e-06, "max_abs_err": 0.0198974609375 }, { "name": "model.layers.31.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 1.0063869012810756e-05, "max_abs_err": 0.0146484375 }, { "name": "model.layers.31.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.57271936890902e-06, "max_abs_err": 0.023681640625 }, { "name": "model.layers.31.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.901929166109767e-06, "max_abs_err": 0.044921875 }, { "name": "model.layers.31.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.851902748574503e-06, "max_abs_err": 0.0390625 }, { "name": "model.layers.31.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.738127689866815e-06, "max_abs_err": 0.051513671875 }, { "name": "model.layers.32.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.459093467332423e-06, "max_abs_err": 0.031494140625 }, { "name": "model.layers.32.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.2684493968845345e-06, "max_abs_err": 0.0177001953125 }, { "name": "model.layers.32.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 1.0188642590946984e-05, "max_abs_err": 0.0164794921875 }, { "name": "model.layers.32.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.540902061009547e-06, "max_abs_err": 0.03125 }, { "name": "model.layers.32.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.763020221318584e-06, "max_abs_err": 0.04931640625 }, { "name": "model.layers.32.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.923499990487471e-06, "max_abs_err": 0.052978515625 }, { "name": "model.layers.32.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.794429959380068e-06, "max_abs_err": 0.048828125 }, { "name": "model.layers.33.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.67969322623685e-06, "max_abs_err": 0.02685546875 }, { "name": "model.layers.33.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.173345925868489e-06, "max_abs_err": 0.0150146484375 }, { "name": "model.layers.33.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 1.134357171395095e-05, "max_abs_err": 0.0172119140625 }, { "name": "model.layers.33.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 8.193464054784272e-06, "max_abs_err": 0.046875 }, { "name": "model.layers.33.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.70071073930012e-06, "max_abs_err": 0.046142578125 }, { "name": "model.layers.33.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 9.019511708174832e-06, "max_abs_err": 0.033447265625 }, { "name": "model.layers.33.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.67576090968214e-06, "max_abs_err": 0.04296875 }, { "name": "model.layers.34.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 7.012407877482474e-06, "max_abs_err": 0.0302734375 }, { "name": "model.layers.34.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 5.800411145173712e-06, "max_abs_err": 0.0242919921875 }, { "name": "model.layers.34.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 1.1260283827141393e-05, "max_abs_err": 0.01708984375 }, { "name": "model.layers.34.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.822169209248386e-06, "max_abs_err": 0.025390625 }, { "name": "model.layers.34.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 7.991255188244395e-06, "max_abs_err": 0.04443359375 }, { "name": "model.layers.34.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.80213610798819e-06, "max_abs_err": 0.038330078125 }, { "name": "model.layers.34.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.149502718879376e-06, "max_abs_err": 0.0537109375 }, { "name": "model.layers.35.self_attn.q_proj", "shape": [ 4096, 2560 ], "dtype": "bfloat16", "mse": 6.9244924816302955e-06, "max_abs_err": 0.0225830078125 }, { "name": "model.layers.35.self_attn.k_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 6.507526450150181e-06, "max_abs_err": 0.04052734375 }, { "name": "model.layers.35.self_attn.v_proj", "shape": [ 1024, 2560 ], "dtype": "bfloat16", "mse": 9.81938501354307e-06, "max_abs_err": 0.0146484375 }, { "name": "model.layers.35.self_attn.o_proj", "shape": [ 2560, 4096 ], "dtype": "bfloat16", "mse": 7.33860088075744e-06, "max_abs_err": 0.037109375 }, { "name": "model.layers.35.mlp.gate_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 8.777114999247715e-06, "max_abs_err": 0.058349609375 }, { "name": "model.layers.35.mlp.up_proj", "shape": [ 9728, 2560 ], "dtype": "bfloat16", "mse": 9.085820238396991e-06, "max_abs_err": 0.05908203125 }, { "name": "model.layers.35.mlp.down_proj", "shape": [ 2560, 9728 ], "dtype": "bfloat16", "mse": 8.085974513960537e-06, "max_abs_err": 0.055419921875 } ], "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." }