| layer,module,loss,samples,damp,time | |
| 0,self_attn.qkv_proj,0.0000029813,0.05000,2.185 | |
| 0,self_attn.o_proj,0.0000001116,0.05000,1.084 | |
| 0,mlp.gate_proj,0.0000072175,0.05000,2.485 | |
| 0,mlp.up_proj,0.0000068269,0.05000,2.472 | |
| 0,mlp.down_proj,0.0000003002,0.05000,2.818 | |
| 1,self_attn.qkv_proj,0.0000198021,0.05000,1.178 | |
| 1,self_attn.o_proj,0.0000000840,0.05000,1.074 | |
| 1,mlp.gate_proj,0.0000090921,0.05000,2.970 | |
| 1,mlp.up_proj,0.0000082698,0.05000,3.030 | |
| 1,mlp.down_proj,0.0000004386,0.05000,3.012 | |
| 2,self_attn.qkv_proj,0.0000369951,0.05000,1.042 | |
| 2,self_attn.o_proj,0.0000001405,0.05000,1.034 | |
| 2,mlp.gate_proj,0.0000112209,0.05000,3.049 | |
| 2,mlp.up_proj,0.0000100619,0.05000,3.136 | |
| 2,mlp.down_proj,0.0000003945,0.05000,2.978 | |
| 3,self_attn.qkv_proj,0.0000493181,0.05000,1.000 | |
| 3,self_attn.o_proj,0.0000001907,0.05000,1.057 | |
| 3,mlp.gate_proj,0.0000116083,0.05000,2.194 | |
| 3,mlp.up_proj,0.0000099847,0.05000,2.211 | |
| 3,mlp.down_proj,0.0001575425,0.05000,2.915 | |
| 4,self_attn.qkv_proj,0.0000429134,0.05000,1.015 | |
| 4,self_attn.o_proj,0.0000005166,0.05000,1.036 | |
| 4,mlp.gate_proj,0.0000153442,0.05000,2.213 | |
| 4,mlp.up_proj,0.0000123836,0.05000,2.234 | |
| 4,mlp.down_proj,0.0000009558,0.05000,2.901 | |
| 5,self_attn.qkv_proj,0.0000511098,0.05000,1.029 | |
| 5,self_attn.o_proj,0.0000007915,0.05000,0.992 | |
| 5,mlp.gate_proj,0.0000157261,0.05000,2.142 | |
| 5,mlp.up_proj,0.0000123270,0.05000,2.167 | |
| 5,mlp.down_proj,0.0000011092,0.05000,2.916 | |
| 6,self_attn.qkv_proj,0.0000368397,0.05000,1.005 | |
| 6,self_attn.o_proj,0.0000010816,0.05000,1.004 | |
| 6,mlp.gate_proj,0.0000144746,0.05000,2.321 | |
| 6,mlp.up_proj,0.0000120766,0.05000,2.343 | |
| 6,mlp.down_proj,0.0000013575,0.05000,2.901 | |
| 7,self_attn.qkv_proj,0.0000558010,0.05000,0.997 | |
| 7,self_attn.o_proj,0.0000015489,0.05000,1.001 | |
| 7,mlp.up_proj,0.0000107158,0.05000,2.283 | |
| 7,mlp.gate_proj,0.0000121614,0.05000,2.292 | |
| 7,mlp.down_proj,0.0000013054,0.05000,2.919 | |
| 8,self_attn.qkv_proj,0.0000478846,0.05000,1.005 | |
| 8,self_attn.o_proj,0.0000010530,0.05000,1.015 | |
| 8,mlp.gate_proj,0.0000116524,0.05000,2.327 | |
| 8,mlp.up_proj,0.0000112374,0.05000,2.347 | |
| 8,mlp.down_proj,0.0000012097,0.05000,2.928 | |
| 9,self_attn.qkv_proj,0.0000377997,0.05000,1.027 | |
| 9,self_attn.o_proj,0.0000008833,0.05000,1.020 | |
| 9,mlp.up_proj,0.0000130112,0.05000,2.346 | |
| 9,mlp.gate_proj,0.0000130494,0.05000,2.366 | |
| 9,mlp.down_proj,0.0000013272,0.05000,2.956 | |
| 10,self_attn.qkv_proj,0.0000416774,0.05000,1.002 | |
| 10,self_attn.o_proj,0.0000014399,0.05000,0.995 | |
| 10,mlp.gate_proj,0.0000147789,0.05000,2.278 | |
| 10,mlp.up_proj,0.0000146073,0.05000,2.298 | |
| 10,mlp.down_proj,0.0000016522,0.05000,2.917 | |
| 11,self_attn.qkv_proj,0.0000451865,0.05000,0.999 | |
| 11,self_attn.o_proj,0.0000025101,0.05000,1.019 | |
| 11,mlp.up_proj,0.0000155388,0.05000,2.297 | |
| 11,mlp.gate_proj,0.0000147828,0.05000,2.319 | |
| 11,mlp.down_proj,0.0000018731,0.05000,2.878 | |
| 12,self_attn.qkv_proj,0.0000411135,0.05000,0.997 | |
| 12,self_attn.o_proj,0.0000016274,0.05000,0.984 | |
| 12,mlp.gate_proj,0.0000138808,0.05000,2.238 | |
| 12,mlp.up_proj,0.0000153072,0.05000,2.255 | |
| 12,mlp.down_proj,0.0000019512,0.05000,2.913 | |
| 13,self_attn.qkv_proj,0.0000524880,0.05000,1.055 | |
| 13,self_attn.o_proj,0.0000025486,0.05000,0.983 | |
| 13,mlp.gate_proj,0.0000134808,0.05000,2.241 | |
| 13,mlp.up_proj,0.0000157018,0.05000,2.267 | |
| 13,mlp.down_proj,0.0000017552,0.05000,2.904 | |
| 14,self_attn.qkv_proj,0.0000434523,0.05000,1.007 | |
| 14,self_attn.o_proj,0.0000018440,0.05000,1.004 | |
| 14,mlp.up_proj,0.0000157787,0.05000,2.369 | |
| 14,mlp.gate_proj,0.0000136520,0.05000,2.391 | |
| 14,mlp.down_proj,0.0000018066,0.05000,2.874 | |
| 15,self_attn.qkv_proj,0.0000482138,0.05000,1.006 | |
| 15,self_attn.o_proj,0.0000023607,0.05000,0.996 | |
| 15,mlp.up_proj,0.0000184480,0.05000,2.258 | |
| 15,mlp.gate_proj,0.0000159979,0.05000,2.285 | |
| 15,mlp.down_proj,0.0000022394,0.05000,2.879 | |
| 16,self_attn.qkv_proj,0.0000460684,0.05000,1.004 | |
| 16,self_attn.o_proj,0.0000029966,0.05000,0.996 | |
| 16,mlp.up_proj,0.0000173775,0.05000,2.229 | |
| 16,mlp.gate_proj,0.0000149928,0.05000,2.250 | |
| 16,mlp.down_proj,0.0000026397,0.05000,2.927 | |
| 17,self_attn.qkv_proj,0.0000391842,0.05000,1.013 | |
| 17,self_attn.o_proj,0.0000019919,0.05000,1.004 | |
| 17,mlp.gate_proj,0.0000157793,0.05000,2.170 | |
| 17,mlp.up_proj,0.0000185717,0.05000,2.195 | |
| 17,mlp.down_proj,0.0000032989,0.05000,2.975 | |
| 18,self_attn.qkv_proj,0.0000400155,0.05000,1.004 | |
| 18,self_attn.o_proj,0.0000047167,0.05000,1.026 | |
| 18,mlp.gate_proj,0.0000173240,0.05000,2.357 | |
| 18,mlp.up_proj,0.0000200385,0.05000,2.378 | |
| 18,mlp.down_proj,0.0000046034,0.05000,2.894 | |
| 19,self_attn.qkv_proj,0.0000434654,0.05000,0.989 | |
| 19,self_attn.o_proj,0.0000067645,0.05000,1.010 | |
| 19,mlp.gate_proj,0.0000196353,0.05000,2.235 | |
| 19,mlp.up_proj,0.0000223742,0.05000,2.258 | |
| 19,mlp.down_proj,0.0000056410,0.05000,2.882 | |
| 20,self_attn.qkv_proj,0.0000463234,0.05000,1.051 | |
| 20,self_attn.o_proj,0.0000086217,0.05000,0.987 | |
| 20,mlp.gate_proj,0.0000240742,0.05000,2.232 | |
| 20,mlp.up_proj,0.0000274313,0.05000,2.254 | |
| 20,mlp.down_proj,0.0000060875,0.05000,2.883 | |
| 21,self_attn.qkv_proj,0.0000501621,0.05000,1.009 | |
| 21,self_attn.o_proj,0.0000087416,0.05000,1.061 | |
| 21,mlp.gate_proj,0.0000260591,0.05000,2.269 | |
| 21,mlp.up_proj,0.0000302359,0.05000,2.288 | |
| 21,mlp.down_proj,0.0000079578,0.05000,2.881 | |
| 22,self_attn.qkv_proj,0.0000489731,0.05000,1.012 | |
| 22,self_attn.o_proj,0.0000178103,0.05000,1.010 | |
| 22,mlp.gate_proj,0.0000266433,0.05000,2.329 | |
| 22,mlp.up_proj,0.0000314974,0.05000,2.348 | |
| 22,mlp.down_proj,0.0000063287,0.05000,2.896 | |
| 23,self_attn.qkv_proj,0.0000455062,0.05000,1.014 | |
| 23,self_attn.o_proj,0.0000040400,0.05000,1.016 | |
| 23,mlp.gate_proj,0.0000260997,0.05000,2.268 | |
| 23,mlp.up_proj,0.0000305928,0.05000,2.296 | |
| 23,mlp.down_proj,0.0000047429,0.05000,2.939 | |
| 24,self_attn.qkv_proj,0.0000446666,0.05000,1.041 | |
| 24,self_attn.o_proj,0.0000039479,0.05000,1.019 | |
| 24,mlp.gate_proj,0.0000253762,0.05000,2.208 | |
| 24,mlp.up_proj,0.0000301786,0.05000,2.233 | |
| 24,mlp.down_proj,0.0000058199,0.05000,2.930 | |
| 25,self_attn.qkv_proj,0.0000491577,0.05000,1.018 | |
| 25,self_attn.o_proj,0.0000154031,0.05000,0.992 | |
| 25,mlp.up_proj,0.0000329190,0.05000,2.297 | |
| 25,mlp.gate_proj,0.0000275557,0.05000,2.325 | |
| 25,mlp.down_proj,0.0000069694,0.05000,2.888 | |
| 26,self_attn.qkv_proj,0.0000472160,0.05000,1.000 | |
| 26,self_attn.o_proj,0.0000240206,0.05000,1.013 | |
| 26,mlp.up_proj,0.0000331026,0.05000,2.370 | |
| 26,mlp.gate_proj,0.0000278213,0.05000,2.383 | |
| 26,mlp.down_proj,0.0000075888,0.05000,2.879 | |
| 27,self_attn.qkv_proj,0.0000390911,0.05000,0.997 | |
| 27,self_attn.o_proj,0.0000069230,0.05000,0.982 | |
| 27,mlp.up_proj,0.0000385140,0.05000,2.292 | |
| 27,mlp.gate_proj,0.0000329811,0.05000,2.326 | |
| 27,mlp.down_proj,0.0000150313,0.05000,2.866 | |
| 28,self_attn.qkv_proj,0.0000598709,0.05000,1.005 | |
| 28,self_attn.o_proj,0.0000301717,0.05000,0.998 | |
| 28,mlp.up_proj,0.0000483479,0.05000,2.217 | |
| 28,mlp.gate_proj,0.0000434695,0.05000,2.222 | |
| 28,mlp.down_proj,0.0000254065,0.05000,2.887 | |
| 29,self_attn.qkv_proj,0.0000556419,0.05000,1.006 | |
| 29,self_attn.o_proj,0.0000174969,0.05000,0.994 | |
| 29,mlp.up_proj,0.0000489981,0.05000,2.362 | |
| 29,mlp.gate_proj,0.0000497381,0.05000,2.376 | |
| 29,mlp.down_proj,0.0000467694,0.05000,2.925 | |
| 30,self_attn.qkv_proj,0.0000398120,0.05000,1.037 | |
| 30,self_attn.o_proj,0.0000189608,0.05000,0.996 | |
| 30,mlp.up_proj,0.0000495070,0.05000,2.266 | |
| 30,mlp.gate_proj,0.0000596360,0.05000,2.280 | |
| 30,mlp.down_proj,0.0003538760,0.05000,2.929 | |
| 31,self_attn.qkv_proj,0.0000012725,0.05000,1.016 | |
| 31,self_attn.o_proj,0.0000023444,0.05000,0.995 | |
| 31,mlp.up_proj,0.0000169833,0.05000,2.230 | |
| 31,mlp.gate_proj,0.0000333658,0.05000,2.260 | |
| 31,mlp.down_proj,0.0001084470,0.05000,2.915 | |