| rank relative_mse precision parameters group members |
| 1 1.644905332612e-03 BF16 41943040 linear::layers.49.self_attn.o_proj layers.49.self_attn.o_proj |
| 2 1.547668504183e-03 BF16 41943040 linear::layers.48.self_attn.o_proj layers.48.self_attn.o_proj |
| 3 1.536447363098e-03 BF16 41943040 linear::layers.42.self_attn.o_proj layers.42.self_attn.o_proj |
| 4 1.532915882594e-03 BF16 41943040 linear::layers.44.self_attn.o_proj layers.44.self_attn.o_proj |
| 5 1.501263897131e-03 BF16 41943040 linear::layers.46.self_attn.o_proj layers.46.self_attn.o_proj |
| 6 1.433183881512e-03 BF16 41943040 linear::layers.47.self_attn.o_proj layers.47.self_attn.o_proj |
| 7 1.320223541732e-03 BF16 41943040 linear::layers.30.self_attn.o_proj layers.30.self_attn.o_proj |
| 8 1.207858975801e-03 BF16 41943040 linear::layers.24.self_attn.o_proj layers.24.self_attn.o_proj |
| 9 1.200786353352e-03 BF16 41943040 linear::layers.43.self_attn.o_proj layers.43.self_attn.o_proj |
| 10 1.175424311008e-03 BF16 41943040 linear::layers.38.self_attn.o_proj layers.38.self_attn.o_proj |
| 11 1.172402569165e-03 BF16 41943040 linear::layers.45.self_attn.o_proj layers.45.self_attn.o_proj |
| 12 1.153195713830e-03 BF16 41943040 linear::layers.35.self_attn.o_proj layers.35.self_attn.o_proj |
| 13 1.142452760433e-03 BF16 131072000 linear::layers.34.mlp.down_proj layers.34.mlp.down_proj |
| 14 1.137696606777e-03 BF16 131072000 linear::layers.26.mlp.down_proj layers.26.mlp.down_proj |
| 15 1.137589244436e-03 BF16 131072000 linear::layers.48.mlp.down_proj layers.48.mlp.down_proj |
| 16 1.116271875272e-03 BF16 131072000 linear::layers.22.mlp.down_proj layers.22.mlp.down_proj |
| 17 1.108265102131e-03 BF16 131072000 linear::layers.29.mlp.down_proj layers.29.mlp.down_proj |
| 18 1.107530456069e-03 BF16 131072000 linear::layers.28.mlp.down_proj layers.28.mlp.down_proj |
| 19 1.100251047260e-03 BF16 41943040 linear::layers.31.self_attn.o_proj layers.31.self_attn.o_proj |
| 20 1.092745538578e-03 BF16 131072000 linear::layers.30.mlp.down_proj layers.30.mlp.down_proj |
| 21 1.090373060838e-03 BF16 131072000 linear::layers.23.mlp.down_proj layers.23.mlp.down_proj |
| 22 1.083811296823e-03 BF16 131072000 linear::layers.24.mlp.down_proj layers.24.mlp.down_proj |
| 23 1.082856470668e-03 BF16 131072000 linear::layers.27.mlp.down_proj layers.27.mlp.down_proj |
| 24 1.075693959445e-03 BF16 131072000 linear::layers.49.mlp.down_proj layers.49.mlp.down_proj |
| 25 1.075439905164e-03 BF16 131072000 linear::layers.36.mlp.down_proj layers.36.mlp.down_proj |
| 26 1.063777452939e-03 BF16 41943040 linear::layers.25.self_attn.o_proj layers.25.self_attn.o_proj |
| 27 1.061575874994e-03 BF16 131072000 linear::layers.17.mlp.down_proj layers.17.mlp.down_proj |
| 28 1.057192779020e-03 BF16 131072000 linear::layers.18.mlp.down_proj layers.18.mlp.down_proj |
| 29 1.054107668707e-03 BF16 131072000 linear::layers.21.mlp.down_proj layers.21.mlp.down_proj |
| 30 1.051651921624e-03 BF16 41943040 linear::layers.37.self_attn.o_proj layers.37.self_attn.o_proj |
| 31 1.050318402853e-03 BF16 131072000 linear::layers.19.mlp.down_proj layers.19.mlp.down_proj |
| 32 1.049361103817e-03 BF16 131072000 linear::layers.46.mlp.down_proj layers.46.mlp.down_proj |
| 33 1.048295120397e-03 BF16 131072000 linear::layers.25.mlp.down_proj layers.25.mlp.down_proj |
| 34 1.047940464856e-03 BF16 41943040 linear::layers.40.self_attn.o_proj layers.40.self_attn.o_proj |
| 35 1.047785761493e-03 FP8 131072000 linear::layers.35.mlp.down_proj layers.35.mlp.down_proj |
| 36 1.036299833157e-03 FP8 41943040 linear::layers.28.self_attn.o_proj layers.28.self_attn.o_proj |
| 37 1.033824137532e-03 FP8 131072000 linear::layers.12.mlp.down_proj layers.12.mlp.down_proj |
| 38 1.021198927451e-03 FP8 131072000 linear::layers.20.mlp.down_proj layers.20.mlp.down_proj |
| 39 1.016771401028e-03 FP8 131072000 linear::layers.41.mlp.down_proj layers.41.mlp.down_proj |
| 40 1.013677642115e-03 FP8 131072000 linear::layers.42.mlp.down_proj layers.42.mlp.down_proj |
| 41 1.013561987395e-03 FP8 131072000 linear::layers.10.mlp.down_proj layers.10.mlp.down_proj |
| 42 1.009103024557e-03 FP8 131072000 linear::layers.15.mlp.down_proj layers.15.mlp.down_proj |
| 43 1.005980795342e-03 FP8 41943040 linear::layers.27.self_attn.o_proj layers.27.self_attn.o_proj |
| 44 9.966944405147e-04 FP8 131072000 linear::layers.39.mlp.down_proj layers.39.mlp.down_proj |
| 45 9.933645384711e-04 FP8 131072000 linear::layers.32.mlp.down_proj layers.32.mlp.down_proj |
| 46 9.859452641074e-04 FP8 131072000 linear::layers.14.mlp.down_proj layers.14.mlp.down_proj |
| 47 9.805422781302e-04 FP8 131072000 linear::layers.11.mlp.down_proj layers.11.mlp.down_proj |
| 48 9.792871861092e-04 FP8 41943040 linear::layers.39.self_attn.o_proj layers.39.self_attn.o_proj |
| 49 9.757749575553e-04 FP8 131072000 linear::layers.16.mlp.down_proj layers.16.mlp.down_proj |
| 50 9.647846049505e-04 FP8 131072000 linear::layers.13.mlp.down_proj layers.13.mlp.down_proj |
| 51 9.631741276790e-04 FP8 131072000 linear::layers.38.mlp.down_proj layers.38.mlp.down_proj |
| 52 9.573445004375e-04 FP8 131072000 linear::layers.37.mlp.down_proj layers.37.mlp.down_proj |
| 53 9.570369106127e-04 FP8 41943040 linear::layers.29.self_attn.o_proj layers.29.self_attn.o_proj |
| 54 9.539916903886e-04 FP8 41943040 linear::layers.36.self_attn.o_proj layers.36.self_attn.o_proj |
| 55 9.331376304085e-04 FP8 131072000 linear::layers.47.mlp.down_proj layers.47.mlp.down_proj |
| 56 9.292229298699e-04 FP8 41943040 linear::layers.34.self_attn.o_proj layers.34.self_attn.o_proj |
| 57 9.213870051549e-04 FP8 41943040 linear::layers.23.self_attn.o_proj layers.23.self_attn.o_proj |
| 58 9.183763761154e-04 FP8 131072000 linear::layers.8.mlp.down_proj layers.8.mlp.down_proj |
| 59 9.016936045446e-04 FP8 131072000 linear::layers.33.mlp.down_proj layers.33.mlp.down_proj |
| 60 9.004733240410e-04 FP8 262144000 gate_up::layers.22.mlp layers.22.mlp.gate_proj,layers.22.mlp.up_proj |
| 61 8.918872858573e-04 FP8 262144000 gate_up::layers.11.mlp layers.11.mlp.gate_proj,layers.11.mlp.up_proj |
| 62 8.914683829096e-04 FP8 41943040 linear::layers.22.self_attn.o_proj layers.22.self_attn.o_proj |
| 63 8.872367525309e-04 FP8 262144000 gate_up::layers.23.mlp layers.23.mlp.gate_proj,layers.23.mlp.up_proj |
| 64 8.841689692781e-04 FP8 262144000 gate_up::layers.10.mlp layers.10.mlp.gate_proj,layers.10.mlp.up_proj |
| 65 8.841446595501e-04 FP8 262144000 gate_up::layers.19.mlp layers.19.mlp.gate_proj,layers.19.mlp.up_proj |
| 66 8.818184128474e-04 FP8 262144000 gate_up::layers.24.mlp layers.24.mlp.gate_proj,layers.24.mlp.up_proj |
| 67 8.812930762932e-04 FP8 41943040 linear::layers.19.self_attn.o_proj layers.19.self_attn.o_proj |
| 68 8.789220954463e-04 FP8 262144000 gate_up::layers.20.mlp layers.20.mlp.gate_proj,layers.20.mlp.up_proj |
| 69 8.765070998812e-04 FP8 41943040 linear::layers.21.self_attn.o_proj layers.21.self_attn.o_proj |
| 70 8.764308989083e-04 FP8 262144000 gate_up::layers.21.mlp layers.21.mlp.gate_proj,layers.21.mlp.up_proj |
| 71 8.744895908167e-04 FP8 262144000 gate_up::layers.34.mlp layers.34.mlp.gate_proj,layers.34.mlp.up_proj |
| 72 8.726309893222e-04 FP8 131072000 linear::layers.40.mlp.down_proj layers.40.mlp.down_proj |
| 73 8.710345733115e-04 FP8 41943040 linear::layers.41.self_attn.o_proj layers.41.self_attn.o_proj |
| 74 8.592773304607e-04 FP8 131072000 linear::layers.5.mlp.down_proj layers.5.mlp.down_proj |
| 75 8.573134497294e-04 FP8 41943040 linear::layers.11.self_attn.o_proj layers.11.self_attn.o_proj |
| 76 8.565416167721e-04 FP8 262144000 gate_up::layers.12.mlp layers.12.mlp.gate_proj,layers.12.mlp.up_proj |
| 77 8.562982364602e-04 FP8 262144000 gate_up::layers.15.mlp layers.15.mlp.gate_proj,layers.15.mlp.up_proj |
| 78 8.531324418050e-04 FP8 41943040 linear::layers.33.self_attn.o_proj layers.33.self_attn.o_proj |
| 79 8.457170623090e-04 FP8 262144000 gate_up::layers.17.mlp layers.17.mlp.gate_proj,layers.17.mlp.up_proj |
| 80 8.447256116782e-04 FP8 262144000 gate_up::layers.25.mlp layers.25.mlp.gate_proj,layers.25.mlp.up_proj |
| 81 8.400224775269e-04 FP8 41943040 linear::layers.26.self_attn.o_proj layers.26.self_attn.o_proj |
| 82 8.368774077547e-04 FP8 262144000 gate_up::layers.26.mlp layers.26.mlp.gate_proj,layers.26.mlp.up_proj |
| 83 8.368199234751e-04 FP8 131072000 linear::layers.44.mlp.down_proj layers.44.mlp.down_proj |
| 84 8.276378265995e-04 FP8 41943040 linear::layers.14.self_attn.o_proj layers.14.self_attn.o_proj |
| 85 8.198584095220e-04 FP8 41943040 linear::layers.32.self_attn.o_proj layers.32.self_attn.o_proj |
| 86 8.194438554151e-04 FP8 262144000 gate_up::layers.18.mlp layers.18.mlp.gate_proj,layers.18.mlp.up_proj |
| 87 8.115857426304e-04 FP8 131072000 linear::layers.31.mlp.down_proj layers.31.mlp.down_proj |
| 88 8.086569052056e-04 FP8 41943040 linear::layers.20.self_attn.o_proj layers.20.self_attn.o_proj |
| 89 8.047650181544e-04 FP8 262144000 gate_up::layers.35.mlp layers.35.mlp.gate_proj,layers.35.mlp.up_proj |
| 90 8.012509107278e-04 FP8 262144000 gate_up::layers.27.mlp layers.27.mlp.gate_proj,layers.27.mlp.up_proj |
| 91 7.964024285771e-04 FP8 262144000 gate_up::layers.14.mlp layers.14.mlp.gate_proj,layers.14.mlp.up_proj |
| 92 7.889474795434e-04 FP8 262144000 gate_up::layers.9.mlp layers.9.mlp.gate_proj,layers.9.mlp.up_proj |
| 93 7.863127552736e-04 FP8 262144000 gate_up::layers.16.mlp layers.16.mlp.gate_proj,layers.16.mlp.up_proj |
| 94 7.846631109890e-04 FP8 131072000 linear::layers.45.mlp.down_proj layers.45.mlp.down_proj |
| 95 7.827324928930e-04 FP8 262144000 gate_up::layers.28.mlp layers.28.mlp.gate_proj,layers.28.mlp.up_proj |
| 96 7.797672501108e-04 FP8 262144000 gate_up::layers.47.mlp layers.47.mlp.gate_proj,layers.47.mlp.up_proj |
| 97 7.710270188577e-04 FP8 262144000 gate_up::layers.46.mlp layers.46.mlp.gate_proj,layers.46.mlp.up_proj |
| 98 7.634876282443e-04 FP8 41943040 linear::layers.13.self_attn.o_proj layers.13.self_attn.o_proj |
| 99 7.614215226548e-04 FP8 262144000 gate_up::layers.48.mlp layers.48.mlp.gate_proj,layers.48.mlp.up_proj |
| 100 7.586616796440e-04 FP8 262144000 gate_up::layers.13.mlp layers.13.mlp.gate_proj,layers.13.mlp.up_proj |
| 101 7.581983119694e-04 FP8 262144000 gate_up::layers.49.mlp layers.49.mlp.gate_proj,layers.49.mlp.up_proj |
| 102 7.465519862330e-04 FP8 262144000 gate_up::layers.43.mlp layers.43.mlp.gate_proj,layers.43.mlp.up_proj |
| 103 7.364132256481e-04 FP8 41943040 linear::layers.18.self_attn.o_proj layers.18.self_attn.o_proj |
| 104 7.326954109690e-04 FP8 262144000 gate_up::layers.44.mlp layers.44.mlp.gate_proj,layers.44.mlp.up_proj |
| 105 7.261024749773e-04 FP8 262144000 gate_up::layers.45.mlp layers.45.mlp.gate_proj,layers.45.mlp.up_proj |
| 106 7.125850763160e-04 FP8 41943040 linear::layers.12.self_attn.o_proj layers.12.self_attn.o_proj |
| 107 7.104821534947e-04 FP8 262144000 gate_up::layers.36.mlp layers.36.mlp.gate_proj,layers.36.mlp.up_proj |
| 108 7.010528021852e-04 FP8 262144000 gate_up::layers.42.mlp layers.42.mlp.gate_proj,layers.42.mlp.up_proj |
| 109 6.984599426641e-04 FP8 262144000 gate_up::layers.33.mlp layers.33.mlp.gate_proj,layers.33.mlp.up_proj |
| 110 6.971777304907e-04 FP8 262144000 gate_up::layers.41.mlp layers.41.mlp.gate_proj,layers.41.mlp.up_proj |
| 111 6.893470420324e-04 FP8 131072000 linear::layers.3.mlp.down_proj layers.3.mlp.down_proj |
| 112 6.872426298753e-04 FP8 41943040 linear::layers.7.self_attn.o_proj layers.7.self_attn.o_proj |
| 113 6.720004899719e-04 FP8 262144000 gate_up::layers.40.mlp layers.40.mlp.gate_proj,layers.40.mlp.up_proj |
| 114 6.715806859788e-04 FP8 262144000 gate_up::layers.39.mlp layers.39.mlp.gate_proj,layers.39.mlp.up_proj |
| 115 6.584419997463e-04 FP8 262144000 gate_up::layers.29.mlp layers.29.mlp.gate_proj,layers.29.mlp.up_proj |
| 116 6.469047188960e-04 FP8 41943040 linear::layers.15.self_attn.o_proj layers.15.self_attn.o_proj |
| 117 6.447572396693e-04 FP8 41943040 linear::layers.9.self_attn.o_proj layers.9.self_attn.o_proj |
| 118 6.358877470629e-04 FP8 41943040 linear::layers.10.self_attn.o_proj layers.10.self_attn.o_proj |
| 119 6.069202529668e-04 FP8 41943040 linear::layers.8.self_attn.o_proj layers.8.self_attn.o_proj |
| 120 6.058075955291e-04 FP8 41943040 linear::layers.16.self_attn.o_proj layers.16.self_attn.o_proj |
| 121 5.954798116377e-04 FP8 262144000 gate_up::layers.38.mlp layers.38.mlp.gate_proj,layers.38.mlp.up_proj |
| 122 5.882689096100e-04 FP8 41943040 linear::layers.3.self_attn.o_proj layers.3.self_attn.o_proj |
| 123 5.789923247235e-04 FP8 131072000 linear::layers.9.mlp.down_proj layers.9.mlp.down_proj |
| 124 5.466728668065e-04 FP8 262144000 gate_up::layers.37.mlp layers.37.mlp.gate_proj,layers.37.mlp.up_proj |
| 125 5.272238257713e-04 FP8 41943040 linear::layers.17.self_attn.o_proj layers.17.self_attn.o_proj |
| 126 5.139021995102e-04 FP8 131072000 linear::layers.2.mlp.down_proj layers.2.mlp.down_proj |
| 127 4.955638721301e-04 FP8 41943040 linear::layers.4.self_attn.o_proj layers.4.self_attn.o_proj |
| 128 4.800037212563e-04 FP8 131072000 linear::layers.4.mlp.down_proj layers.4.mlp.down_proj |
| 129 4.647992739660e-04 FP8 41943040 linear::layers.2.self_attn.o_proj layers.2.self_attn.o_proj |
| 130 4.103554211562e-04 FP8 41943040 linear::layers.5.self_attn.o_proj layers.5.self_attn.o_proj |
| 131 4.040372681756e-04 FP8 41943040 linear::layers.6.self_attn.o_proj layers.6.self_attn.o_proj |
| 132 4.015835952462e-04 FP8 131072000 linear::layers.1.mlp.down_proj layers.1.mlp.down_proj |
| 133 3.936916902321e-04 FP8 52428800 qkv::layers.19.self_attn.main layers.19.self_attn.k_proj,layers.19.self_attn.q_proj,layers.19.self_attn.v_proj |
| 134 3.893350167008e-04 FP8 262144000 gate_up::layers.0.mlp layers.0.mlp.gate_proj,layers.0.mlp.up_proj |
| 135 3.575691188401e-04 FP8 52428800 qkv::layers.20.self_attn.main layers.20.self_attn.k_proj,layers.20.self_attn.q_proj,layers.20.self_attn.v_proj |
| 136 3.553887298200e-04 FP8 52428800 qkv::layers.18.self_attn.main layers.18.self_attn.k_proj,layers.18.self_attn.q_proj,layers.18.self_attn.v_proj |
| 137 3.502794627363e-04 FP8 52428800 qkv::layers.25.self_attn.main layers.25.self_attn.k_proj,layers.25.self_attn.q_proj,layers.25.self_attn.v_proj |
| 138 3.369211208493e-04 FP8 52428800 qkv::layers.23.self_attn.main layers.23.self_attn.k_proj,layers.23.self_attn.q_proj,layers.23.self_attn.v_proj |
| 139 3.368198495095e-04 FP8 52428800 qkv::layers.12.self_attn.main layers.12.self_attn.k_proj,layers.12.self_attn.q_proj,layers.12.self_attn.v_proj |
| 140 3.304155883933e-04 FP8 52428800 qkv::layers.47.self_attn.main layers.47.self_attn.k_proj,layers.47.self_attn.q_proj,layers.47.self_attn.v_proj |
| 141 3.267026805022e-04 FP8 52428800 qkv::layers.17.self_attn.main layers.17.self_attn.k_proj,layers.17.self_attn.q_proj,layers.17.self_attn.v_proj |
| 142 3.229925413466e-04 FP8 52428800 qkv::layers.22.self_attn.main layers.22.self_attn.k_proj,layers.22.self_attn.q_proj,layers.22.self_attn.v_proj |
| 143 3.207959214506e-04 FP8 52428800 qkv::layers.32.self_attn.main layers.32.self_attn.k_proj,layers.32.self_attn.q_proj,layers.32.self_attn.v_proj |
| 144 3.197798556462e-04 FP8 52428800 qkv::layers.9.self_attn.main layers.9.self_attn.k_proj,layers.9.self_attn.q_proj,layers.9.self_attn.v_proj |
| 145 3.169524757152e-04 FP8 52428800 qkv::layers.21.self_attn.main layers.21.self_attn.k_proj,layers.21.self_attn.q_proj,layers.21.self_attn.v_proj |
| 146 3.151848694664e-04 FP8 52428800 qkv::layers.29.self_attn.main layers.29.self_attn.k_proj,layers.29.self_attn.q_proj,layers.29.self_attn.v_proj |
| 147 3.146310562469e-04 FP8 52428800 qkv::layers.10.self_attn.main layers.10.self_attn.k_proj,layers.10.self_attn.q_proj,layers.10.self_attn.v_proj |
| 148 3.134114650614e-04 FP8 52428800 qkv::layers.31.self_attn.main layers.31.self_attn.k_proj,layers.31.self_attn.q_proj,layers.31.self_attn.v_proj |
| 149 3.126033762219e-04 FP8 52428800 qkv::layers.13.self_attn.main layers.13.self_attn.k_proj,layers.13.self_attn.q_proj,layers.13.self_attn.v_proj |
| 150 3.107748270228e-04 FP8 52428800 qkv::layers.11.self_attn.main layers.11.self_attn.k_proj,layers.11.self_attn.q_proj,layers.11.self_attn.v_proj |
| 151 3.081961344141e-04 FP8 52428800 qkv::layers.49.self_attn.main layers.49.self_attn.k_proj,layers.49.self_attn.q_proj,layers.49.self_attn.v_proj |
| 152 3.076858304540e-04 FP8 52428800 qkv::layers.15.self_attn.main layers.15.self_attn.k_proj,layers.15.self_attn.q_proj,layers.15.self_attn.v_proj |
| 153 3.051550972768e-04 FP8 41943040 linear::layers.1.self_attn.o_proj layers.1.self_attn.o_proj |
| 154 3.024745300822e-04 FP8 52428800 qkv::layers.14.self_attn.main layers.14.self_attn.k_proj,layers.14.self_attn.q_proj,layers.14.self_attn.v_proj |
| 155 2.990808293538e-04 FP8 52428800 qkv::layers.34.self_attn.main layers.34.self_attn.k_proj,layers.34.self_attn.q_proj,layers.34.self_attn.v_proj |
| 156 2.957962080721e-04 FP8 52428800 qkv::layers.30.self_attn.main layers.30.self_attn.k_proj,layers.30.self_attn.q_proj,layers.30.self_attn.v_proj |
| 157 2.933652692348e-04 FP8 262144000 gate_up::layers.30.mlp layers.30.mlp.gate_proj,layers.30.mlp.up_proj |
| 158 2.917757210709e-04 FP8 52428800 qkv::layers.33.self_attn.main layers.33.self_attn.k_proj,layers.33.self_attn.q_proj,layers.33.self_attn.v_proj |
| 159 2.891928540072e-04 FP8 52428800 qkv::layers.24.self_attn.main layers.24.self_attn.k_proj,layers.24.self_attn.q_proj,layers.24.self_attn.v_proj |
| 160 2.883680266905e-04 FP8 52428800 qkv::layers.16.self_attn.main layers.16.self_attn.k_proj,layers.16.self_attn.q_proj,layers.16.self_attn.v_proj |
| 161 2.882303919222e-04 FP8 52428800 qkv::layers.26.self_attn.main layers.26.self_attn.k_proj,layers.26.self_attn.q_proj,layers.26.self_attn.v_proj |
| 162 2.875280205188e-04 FP8 52428800 qkv::layers.27.self_attn.main layers.27.self_attn.k_proj,layers.27.self_attn.q_proj,layers.27.self_attn.v_proj |
| 163 2.798775837751e-04 FP8 52428800 qkv::layers.44.self_attn.main layers.44.self_attn.k_proj,layers.44.self_attn.q_proj,layers.44.self_attn.v_proj |
| 164 2.798056919535e-04 FP8 52428800 qkv::layers.35.self_attn.main layers.35.self_attn.k_proj,layers.35.self_attn.q_proj,layers.35.self_attn.v_proj |
| 165 2.795507868551e-04 FP8 52428800 qkv::layers.45.self_attn.main layers.45.self_attn.k_proj,layers.45.self_attn.q_proj,layers.45.self_attn.v_proj |
| 166 2.790192945110e-04 FP8 52428800 qkv::layers.46.self_attn.main layers.46.self_attn.k_proj,layers.46.self_attn.q_proj,layers.46.self_attn.v_proj |
| 167 2.783886196504e-04 FP8 262144000 gate_up::layers.32.mlp layers.32.mlp.gate_proj,layers.32.mlp.up_proj |
| 168 2.762054071432e-04 FP8 52428800 qkv::layers.8.self_attn.main layers.8.self_attn.k_proj,layers.8.self_attn.q_proj,layers.8.self_attn.v_proj |
| 169 2.745048889001e-04 FP8 41943040 linear::layers.0.self_attn.o_proj layers.0.self_attn.o_proj |
| 170 2.711100120041e-04 FP8 52428800 qkv::layers.43.self_attn.main layers.43.self_attn.k_proj,layers.43.self_attn.q_proj,layers.43.self_attn.v_proj |
| 171 2.706422696729e-04 FP8 52428800 qkv::layers.28.self_attn.main layers.28.self_attn.k_proj,layers.28.self_attn.q_proj,layers.28.self_attn.v_proj |
| 172 2.670172606247e-04 FP8 52428800 qkv::layers.7.self_attn.main layers.7.self_attn.k_proj,layers.7.self_attn.q_proj,layers.7.self_attn.v_proj |
| 173 2.661823828331e-04 FP8 52428800 qkv::layers.42.self_attn.main layers.42.self_attn.k_proj,layers.42.self_attn.q_proj,layers.42.self_attn.v_proj |
| 174 2.631146943054e-04 FP8 131072000 linear::layers.0.mlp.down_proj layers.0.mlp.down_proj |
| 175 2.623688125040e-04 FP8 262144000 gate_up::layers.8.mlp layers.8.mlp.gate_proj,layers.8.mlp.up_proj |
| 176 2.552596957933e-04 FP8 52428800 qkv::layers.48.self_attn.main layers.48.self_attn.k_proj,layers.48.self_attn.q_proj,layers.48.self_attn.v_proj |
| 177 2.508625448609e-04 FP8 52428800 qkv::layers.36.self_attn.main layers.36.self_attn.k_proj,layers.36.self_attn.q_proj,layers.36.self_attn.v_proj |
| 178 2.488009026893e-04 FP8 262144000 gate_up::layers.31.mlp layers.31.mlp.gate_proj,layers.31.mlp.up_proj |
| 179 2.424460407928e-04 FP8 52428800 qkv::layers.41.self_attn.main layers.41.self_attn.k_proj,layers.41.self_attn.q_proj,layers.41.self_attn.v_proj |
| 180 2.419749183877e-04 FP8 52428800 qkv::layers.6.self_attn.main layers.6.self_attn.k_proj,layers.6.self_attn.q_proj,layers.6.self_attn.v_proj |
| 181 2.345472433874e-04 FP8 52428800 qkv::layers.4.self_attn.main layers.4.self_attn.k_proj,layers.4.self_attn.q_proj,layers.4.self_attn.v_proj |
| 182 2.246919581644e-04 FP8 52428800 qkv::layers.38.self_attn.main layers.38.self_attn.k_proj,layers.38.self_attn.q_proj,layers.38.self_attn.v_proj |
| 183 2.216211862301e-04 FP8 52428800 qkv::layers.5.self_attn.main layers.5.self_attn.k_proj,layers.5.self_attn.q_proj,layers.5.self_attn.v_proj |
| 184 2.192777217420e-04 FP8 52428800 qkv::layers.3.self_attn.main layers.3.self_attn.k_proj,layers.3.self_attn.q_proj,layers.3.self_attn.v_proj |
| 185 2.186818859286e-04 FP8 52428800 qkv::layers.39.self_attn.main layers.39.self_attn.k_proj,layers.39.self_attn.q_proj,layers.39.self_attn.v_proj |
| 186 2.152295015065e-04 FP8 52428800 qkv::layers.40.self_attn.main layers.40.self_attn.k_proj,layers.40.self_attn.q_proj,layers.40.self_attn.v_proj |
| 187 2.103234801555e-04 FP8 52428800 qkv::layers.37.self_attn.main layers.37.self_attn.k_proj,layers.37.self_attn.q_proj,layers.37.self_attn.v_proj |
| 188 2.085442845094e-04 FP8 131072000 linear::layers.43.mlp.down_proj layers.43.mlp.down_proj |
| 189 2.050894607884e-04 FP8 52428800 qkv::layers.2.self_attn.main layers.2.self_attn.k_proj,layers.2.self_attn.q_proj,layers.2.self_attn.v_proj |
| 190 1.921664857877e-04 FP8 131072000 linear::layers.7.mlp.down_proj layers.7.mlp.down_proj |
| 191 1.828258347529e-04 FP8 52428800 qkv::layers.1.self_attn.main layers.1.self_attn.k_proj,layers.1.self_attn.q_proj,layers.1.self_attn.v_proj |
| 192 1.480617098853e-04 FP8 52428800 qkv::layers.0.self_attn.main layers.0.self_attn.k_proj,layers.0.self_attn.q_proj,layers.0.self_attn.v_proj |
| 193 1.269788197201e-04 FP8 262144000 gate_up::layers.7.mlp layers.7.mlp.gate_proj,layers.7.mlp.up_proj |
| 194 1.110743596744e-04 FP8 262144000 gate_up::layers.6.mlp layers.6.mlp.gate_proj,layers.6.mlp.up_proj |
| 195 1.011459830825e-04 FP8 262144000 gate_up::layers.5.mlp layers.5.mlp.gate_proj,layers.5.mlp.up_proj |
| 196 7.527054696695e-05 FP8 262144000 gate_up::layers.3.mlp layers.3.mlp.gate_proj,layers.3.mlp.up_proj |
| 197 7.380306649845e-05 FP8 262144000 gate_up::layers.2.mlp layers.2.mlp.gate_proj,layers.2.mlp.up_proj |
| 198 6.955453561496e-05 FP8 262144000 gate_up::layers.4.mlp layers.4.mlp.gate_proj,layers.4.mlp.up_proj |
| 199 6.855085193762e-05 FP8 262144000 gate_up::layers.1.mlp layers.1.mlp.gate_proj,layers.1.mlp.up_proj |
| 200 4.425255642320e-05 FP8 131072000 linear::layers.6.mlp.down_proj layers.6.mlp.down_proj |
|
|