|
layer,module,loss,samples,damp,time
|
|
0,self_attn.k_proj,0.0000000144,0.01000,2.929
|
|
0,self_attn.v_proj,0.0000000149,0.01000,1.526
|
|
0,self_attn.q_proj,0.0000003634,0.01000,1.537
|
|
0,self_attn.o_proj,0.0001249238,0.01000,1.510
|
|
0,mlp.gate_up_proj,0.0000759352,0.01000,1.753
|
|
0,mlp.down_proj,4.9248185034,0.01000,7.646
|
|
1,self_attn.k_proj,0.0000166166,0.01000,1.470
|
|
1,self_attn.v_proj,0.0000085691,0.01000,1.496
|
|
1,self_attn.q_proj,0.0000438481,0.01000,1.511
|
|
1,self_attn.o_proj,0.6955548700,0.01000,1.495
|
|
1,mlp.gate_up_proj,0.0000933231,0.01000,1.736
|
|
1,mlp.down_proj,0.0001679857,0.01000,7.665
|
|
2,self_attn.k_proj,0.0000002660,0.01000,1.464
|
|
2,self_attn.v_proj,0.0000003256,0.01000,1.489
|
|
2,self_attn.q_proj,0.0000034409,0.01000,1.507
|
|
2,self_attn.o_proj,0.0000137586,0.01000,1.494
|
|
2,mlp.gate_up_proj,0.0001726024,0.01000,1.738
|
|
2,mlp.down_proj,0.0003692205,0.01000,7.631
|
|
3,self_attn.k_proj,0.0000004319,0.01000,1.468
|
|
3,self_attn.v_proj,0.0000004455,0.01000,1.494
|
|
3,self_attn.q_proj,0.0000043855,0.01000,1.515
|
|
3,self_attn.o_proj,0.0000306009,0.01000,1.495
|
|
3,mlp.gate_up_proj,0.0001733597,0.01000,1.763
|
|
3,mlp.down_proj,0.0001968489,0.01000,7.650
|
|
4,self_attn.k_proj,0.0000004558,0.01000,1.472
|
|
4,self_attn.v_proj,0.0000003785,0.01000,1.492
|
|
4,self_attn.q_proj,0.0000057011,0.01000,1.520
|
|
4,self_attn.o_proj,0.0000150201,0.01000,1.497
|
|
4,mlp.gate_up_proj,0.0002348814,0.01000,1.757
|
|
4,mlp.down_proj,0.0001950488,0.01000,7.644
|
|
5,self_attn.k_proj,0.0000007762,0.01000,1.486
|
|
5,self_attn.v_proj,0.0000006402,0.01000,1.513
|
|
5,self_attn.q_proj,0.0000093449,0.01000,1.535
|
|
5,self_attn.o_proj,0.0000127081,0.01000,1.519
|
|
5,mlp.gate_up_proj,0.0004476085,0.01000,1.776
|
|
5,mlp.down_proj,0.0002605490,0.01000,7.720
|
|
6,self_attn.k_proj,0.0000010243,0.01000,1.478
|
|
6,self_attn.v_proj,0.0000008929,0.01000,1.504
|
|
6,self_attn.q_proj,0.0000120158,0.01000,1.522
|
|
6,self_attn.o_proj,0.0000048679,0.01000,1.512
|
|
6,mlp.gate_up_proj,0.0004828231,0.01000,1.774
|
|
6,mlp.down_proj,0.0002191994,0.01000,7.708
|
|
7,self_attn.k_proj,0.0000009690,0.01000,1.513
|
|
7,self_attn.v_proj,0.0000008925,0.01000,1.516
|
|
7,self_attn.q_proj,0.0000119501,0.01000,1.528
|
|
7,self_attn.o_proj,0.0000063818,0.01000,1.517
|
|
7,mlp.gate_up_proj,0.0005537864,0.01000,1.761
|
|
7,mlp.down_proj,0.0002386200,0.01000,7.716
|
|
8,self_attn.k_proj,0.0000013532,0.01000,1.484
|
|
8,self_attn.v_proj,0.0000010706,0.01000,1.516
|
|
8,self_attn.q_proj,0.0000189551,0.01000,1.548
|
|
8,self_attn.o_proj,0.0000090043,0.01000,1.511
|
|
8,mlp.gate_up_proj,0.0005882618,0.01000,1.758
|
|
8,mlp.down_proj,0.0002215634,0.01000,7.721
|
|
9,self_attn.k_proj,0.0000015445,0.01000,1.475
|
|
9,self_attn.v_proj,0.0000011883,0.01000,1.508
|
|
9,self_attn.q_proj,0.0000195079,0.01000,1.534
|
|
9,self_attn.o_proj,0.0000100017,0.01000,1.519
|
|
9,mlp.gate_up_proj,0.0005764399,0.01000,1.756
|
|
9,mlp.down_proj,0.0002461467,0.01000,7.688
|
|
10,self_attn.k_proj,0.0000012949,0.01000,1.484
|
|
10,self_attn.v_proj,0.0000012752,0.01000,1.510
|
|
10,self_attn.q_proj,0.0000177911,0.01000,1.524
|
|
10,self_attn.o_proj,0.0000069640,0.01000,1.512
|
|
10,mlp.gate_up_proj,0.0006654446,0.01000,1.757
|
|
10,mlp.down_proj,0.0002258848,0.01000,7.713
|
|
11,self_attn.k_proj,0.0000015689,0.01000,1.485
|
|
11,self_attn.v_proj,0.0000013155,0.01000,1.518
|
|
11,self_attn.q_proj,0.0000224711,0.01000,1.542
|
|
11,self_attn.o_proj,0.0000176500,0.01000,1.514
|
|
11,mlp.gate_up_proj,0.0005532406,0.01000,1.758
|
|
11,mlp.down_proj,0.0002012330,0.01000,7.737
|
|
12,self_attn.k_proj,0.0000019306,0.01000,1.484
|
|
12,self_attn.v_proj,0.0000012937,0.01000,1.514
|
|
12,self_attn.q_proj,0.0000255264,0.01000,1.570
|
|
12,self_attn.o_proj,0.0000130199,0.01000,1.513
|
|
12,mlp.gate_up_proj,0.0004080150,0.01000,1.756
|
|
12,mlp.down_proj,0.0001199784,0.01000,7.747
|
|
13,self_attn.k_proj,0.0000016401,0.01000,1.471
|
|
13,self_attn.v_proj,0.0000015008,0.01000,1.494
|
|
13,self_attn.q_proj,0.0000262780,0.01000,1.512
|
|
13,self_attn.o_proj,0.0000124686,0.01000,1.498
|
|
13,mlp.gate_up_proj,0.0003443256,0.01000,1.752
|
|
13,mlp.down_proj,0.0000911502,0.01000,7.650
|
|
14,self_attn.k_proj,0.0000019886,0.01000,1.474
|
|
14,self_attn.v_proj,0.0000016479,0.01000,1.509
|
|
14,self_attn.q_proj,0.0000326347,0.01000,1.519
|
|
14,self_attn.o_proj,0.0000130200,0.01000,1.510
|
|
14,mlp.gate_up_proj,0.0002595201,0.01000,1.760
|
|
14,mlp.down_proj,0.0000595815,0.01000,7.650
|
|
15,self_attn.k_proj,0.0000011081,0.01000,1.476
|
|
15,self_attn.v_proj,0.0000010541,0.01000,1.504
|
|
15,self_attn.q_proj,0.0000183640,0.01000,1.575
|
|
15,self_attn.o_proj,0.0000130012,0.01000,1.498
|
|
15,mlp.gate_up_proj,0.0003749305,0.01000,1.745
|
|
15,mlp.down_proj,0.0001039048,0.01000,7.664
|
|
16,self_attn.k_proj,0.0000019731,0.01000,1.486
|
|
16,self_attn.v_proj,0.0000017980,0.01000,1.510
|
|
16,self_attn.q_proj,0.0000333973,0.01000,1.530
|
|
16,self_attn.o_proj,0.0000107102,0.01000,1.514
|
|
16,mlp.gate_up_proj,0.0004214915,0.01000,1.753
|
|
16,mlp.down_proj,0.0001009013,0.01000,7.693
|
|
17,self_attn.k_proj,0.0000016455,0.01000,1.489
|
|
17,self_attn.v_proj,0.0000015577,0.01000,1.521
|
|
17,self_attn.q_proj,0.0000259446,0.01000,1.543
|
|
17,self_attn.o_proj,0.0000083092,0.01000,1.516
|
|
17,mlp.gate_up_proj,0.0003165086,0.01000,1.763
|
|
17,mlp.down_proj,0.0000607031,0.01000,7.700
|
|
18,self_attn.k_proj,0.0000020781,0.01000,1.476
|
|
18,self_attn.v_proj,0.0000018589,0.01000,1.507
|
|
18,self_attn.q_proj,0.0000341080,0.01000,1.532
|
|
18,self_attn.o_proj,0.0000055664,0.01000,1.497
|
|
18,mlp.gate_up_proj,0.0004160402,0.01000,1.759
|
|
18,mlp.down_proj,0.0000883801,0.01000,7.659
|
|
19,self_attn.k_proj,0.0000016065,0.01000,1.477
|
|
19,self_attn.v_proj,0.0000014296,0.01000,1.498
|
|
19,self_attn.q_proj,0.0000263949,0.01000,1.508
|
|
19,self_attn.o_proj,0.0000075932,0.01000,1.498
|
|
19,mlp.gate_up_proj,0.0002235646,0.01000,1.757
|
|
19,mlp.down_proj,0.0000297038,0.01000,7.687
|
|
20,self_attn.k_proj,0.0000018283,0.01000,1.484
|
|
20,self_attn.v_proj,0.0000016944,0.01000,1.516
|
|
20,self_attn.q_proj,0.0000298715,0.01000,1.529
|
|
20,self_attn.o_proj,0.0000067423,0.01000,1.527
|
|
20,mlp.gate_up_proj,0.0003179047,0.01000,1.757
|
|
20,mlp.down_proj,0.0000497135,0.01000,7.741
|
|
21,self_attn.k_proj,0.0000012393,0.01000,1.496
|
|
21,self_attn.v_proj,0.0000014096,0.01000,1.529
|
|
21,self_attn.q_proj,0.0000161398,0.01000,1.562
|
|
21,self_attn.o_proj,0.0000086058,0.01000,1.520
|
|
21,mlp.gate_up_proj,0.0004175537,0.01000,1.765
|
|
21,mlp.down_proj,0.0000624529,0.01000,7.735
|
|
22,self_attn.k_proj,0.0000019124,0.01000,1.488
|
|
22,self_attn.v_proj,0.0000016545,0.01000,1.514
|
|
22,self_attn.q_proj,0.0000287724,0.01000,1.529
|
|
22,self_attn.o_proj,0.0000045201,0.01000,1.519
|
|
22,mlp.gate_up_proj,0.0003837140,0.01000,1.759
|
|
22,mlp.down_proj,0.0000529506,0.01000,7.719
|
|
23,self_attn.k_proj,0.0000017570,0.01000,1.494
|
|
23,self_attn.v_proj,0.0000016161,0.01000,1.520
|
|
23,self_attn.q_proj,0.0000270148,0.01000,1.550
|
|
23,self_attn.o_proj,0.0000016576,0.01000,1.524
|
|
23,mlp.gate_up_proj,0.0004301090,0.01000,1.760
|
|
23,mlp.down_proj,0.0000556484,0.01000,7.735
|
|
24,self_attn.k_proj,0.0000016560,0.01000,1.496
|
|
24,self_attn.v_proj,0.0000021923,0.01000,1.521
|
|
24,self_attn.q_proj,0.0000179768,0.01000,1.544
|
|
24,self_attn.o_proj,0.0000042039,0.01000,1.523
|
|
24,mlp.gate_up_proj,0.0003848278,0.01000,1.782
|
|
24,mlp.down_proj,0.0000444143,0.01000,7.769
|
|
25,self_attn.k_proj,0.0000020776,0.01000,1.499
|
|
25,self_attn.v_proj,0.0000018399,0.01000,1.518
|
|
25,self_attn.q_proj,0.0000304697,0.01000,1.535
|
|
25,self_attn.o_proj,0.0000057855,0.01000,1.560
|
|
25,mlp.gate_up_proj,0.0003310795,0.01000,1.787
|
|
25,mlp.down_proj,0.0000374100,0.01000,7.721
|
|
26,self_attn.k_proj,0.0000021007,0.01000,1.501
|
|
26,self_attn.v_proj,0.0000018366,0.01000,1.553
|
|
26,self_attn.q_proj,0.0000314466,0.01000,1.560
|
|
26,self_attn.o_proj,0.0000047819,0.01000,1.520
|
|
26,mlp.gate_up_proj,0.0003340909,0.01000,1.763
|
|
26,mlp.down_proj,0.0000360316,0.01000,7.740
|
|
27,self_attn.k_proj,0.0000021561,0.01000,1.495
|
|
27,self_attn.v_proj,0.0000018368,0.01000,1.526
|
|
27,self_attn.q_proj,0.0000330897,0.01000,1.543
|
|
27,self_attn.o_proj,0.0000045089,0.01000,1.522
|
|
27,mlp.gate_up_proj,0.0002663189,0.01000,1.770
|
|
27,mlp.down_proj,0.0000259201,0.01000,7.749
|
|
28,self_attn.k_proj,0.0000026158,0.01000,1.498
|
|
28,self_attn.v_proj,0.0000022798,0.01000,1.520
|
|
28,self_attn.q_proj,0.0000386177,0.01000,1.552
|
|
28,self_attn.o_proj,0.0000052657,0.01000,1.522
|
|
28,mlp.gate_up_proj,0.0003065639,0.01000,1.794
|
|
28,mlp.down_proj,0.0000314570,0.01000,7.720
|
|
29,self_attn.k_proj,0.0000023923,0.01000,1.489
|
|
29,self_attn.v_proj,0.0000022201,0.01000,1.515
|
|
29,self_attn.q_proj,0.0000375617,0.01000,1.533
|
|
29,self_attn.o_proj,0.0000060325,0.01000,1.526
|
|
29,mlp.gate_up_proj,0.0002587825,0.01000,1.764
|
|
29,mlp.down_proj,0.0000264966,0.01000,7.725
|
|
30,self_attn.k_proj,0.0000029710,0.01000,1.491
|
|
30,self_attn.v_proj,0.0000028491,0.01000,1.541
|
|
30,self_attn.q_proj,0.0000480602,0.01000,1.540
|
|
30,self_attn.o_proj,0.0000023055,0.01000,1.523
|
|
30,mlp.gate_up_proj,0.0002376178,0.01000,1.758
|
|
30,mlp.down_proj,0.0000210452,0.01000,7.753
|
|
31,self_attn.k_proj,0.0000027275,0.01000,1.496
|
|
31,self_attn.v_proj,0.0000028166,0.01000,1.516
|
|
31,self_attn.q_proj,0.0000448323,0.01000,1.542
|
|
31,self_attn.o_proj,0.0000031633,0.01000,1.537
|
|
31,mlp.gate_up_proj,0.0002244233,0.01000,1.758
|
|
31,mlp.down_proj,0.0000184050,0.01000,7.731
|
|
32,self_attn.k_proj,0.0000028851,0.01000,1.491
|
|
32,self_attn.v_proj,0.0000028860,0.01000,1.524
|
|
32,self_attn.q_proj,0.0000488739,0.01000,1.544
|
|
32,self_attn.o_proj,0.0000038245,0.01000,1.522
|
|
32,mlp.gate_up_proj,0.0002103205,0.01000,1.765
|
|
32,mlp.down_proj,0.0000162122,0.01000,7.755
|
|
33,self_attn.k_proj,0.0000018585,0.01000,1.493
|
|
33,self_attn.v_proj,0.0000018939,0.01000,1.534
|
|
33,self_attn.q_proj,0.0000318633,0.01000,1.554
|
|
33,self_attn.o_proj,0.0000035201,0.01000,1.554
|
|
33,mlp.gate_up_proj,0.0001825548,0.01000,1.782
|
|
33,mlp.down_proj,0.0000126821,0.01000,7.742
|
|
34,self_attn.k_proj,0.0000019409,0.01000,1.504
|
|
34,self_attn.v_proj,0.0000017755,0.01000,1.529
|
|
34,self_attn.q_proj,0.0000312464,0.01000,1.569
|
|
34,self_attn.o_proj,0.0000106473,0.01000,1.523
|
|
34,mlp.gate_up_proj,0.0001308567,0.01000,1.761
|
|
34,mlp.down_proj,0.0000084482,0.01000,7.757
|
|
35,self_attn.k_proj,0.0000020172,0.01000,1.502
|
|
35,self_attn.v_proj,0.0000017908,0.01000,1.526
|
|
35,self_attn.q_proj,0.0000337882,0.01000,1.550
|
|
35,self_attn.o_proj,0.0000087794,0.01000,1.528
|
|
35,mlp.gate_up_proj,0.0001125164,0.01000,1.797
|
|
35,mlp.down_proj,0.0000068781,0.01000,7.763
|
|
36,self_attn.k_proj,0.0000021642,0.01000,1.475
|
|
36,self_attn.v_proj,0.0000020408,0.01000,1.504
|
|
36,self_attn.q_proj,0.0000388233,0.01000,1.540
|
|
36,self_attn.o_proj,0.0000083437,0.01000,1.503
|
|
36,mlp.gate_up_proj,0.0001143253,0.01000,1.747
|
|
36,mlp.down_proj,0.0000070406,0.01000,7.668
|
|
37,self_attn.k_proj,0.0000018855,0.01000,1.474
|
|
37,self_attn.v_proj,0.0000017017,0.01000,1.498
|
|
37,self_attn.q_proj,0.0000330188,0.01000,1.517
|
|
37,self_attn.o_proj,0.0000095858,0.01000,1.521
|
|
37,mlp.gate_up_proj,0.0000955174,0.01000,1.788
|
|
37,mlp.down_proj,0.0000051665,0.01000,7.658
|
|
38,self_attn.k_proj,0.0000026216,0.01000,1.474
|
|
38,self_attn.v_proj,0.0000024581,0.01000,1.500
|
|
38,self_attn.q_proj,0.0000449249,0.01000,1.515
|
|
38,self_attn.o_proj,0.0000140097,0.01000,1.498
|
|
38,mlp.gate_up_proj,0.0000959132,0.01000,1.748
|
|
38,mlp.down_proj,0.0000055332,0.01000,7.754
|
|
39,self_attn.k_proj,0.0000023764,0.01000,1.505
|
|
39,self_attn.v_proj,0.0000021691,0.01000,1.514
|
|
39,self_attn.q_proj,0.0000410026,0.01000,1.538
|
|
39,self_attn.o_proj,0.0000092969,0.01000,1.514
|
|
39,mlp.gate_up_proj,0.0001086052,0.01000,1.762
|
|
39,mlp.down_proj,0.0000062391,0.01000,7.746
|
|
40,self_attn.k_proj,0.0000026411,0.01000,1.513
|
|
40,self_attn.v_proj,0.0000024685,0.01000,1.507
|
|
40,self_attn.q_proj,0.0000463585,0.01000,1.522
|
|
40,self_attn.o_proj,0.0000110130,0.01000,1.514
|
|
40,mlp.gate_up_proj,0.0000871354,0.01000,1.754
|
|
40,mlp.down_proj,0.0000039779,0.01000,7.759
|
|
41,self_attn.k_proj,0.0000020181,0.01000,1.486
|
|
41,self_attn.v_proj,0.0000021271,0.01000,1.507
|
|
41,self_attn.q_proj,0.0000409969,0.01000,1.547
|
|
41,self_attn.o_proj,0.0000054175,0.01000,1.539
|
|
41,mlp.gate_up_proj,0.0000830821,0.01000,1.766
|
|
41,mlp.down_proj,0.0000035916,0.01000,7.751
|
|
42,self_attn.k_proj,0.0000024863,0.01000,1.483
|
|
42,self_attn.v_proj,0.0000025053,0.01000,1.510
|
|
42,self_attn.q_proj,0.0000468063,0.01000,1.528
|
|
42,self_attn.o_proj,0.0000105437,0.01000,1.514
|
|
42,mlp.gate_up_proj,0.0000811200,0.01000,1.754
|
|
42,mlp.down_proj,0.0000035646,0.01000,7.721
|
|
43,self_attn.k_proj,0.0000023590,0.01000,1.483
|
|
43,self_attn.v_proj,0.0000022048,0.01000,1.507
|
|
43,self_attn.q_proj,0.0000429752,0.01000,1.530
|
|
43,self_attn.o_proj,0.0000107138,0.01000,1.521
|
|
43,mlp.gate_up_proj,0.0000761639,0.01000,1.764
|
|
43,mlp.down_proj,0.0000033740,0.01000,7.753
|
|
44,self_attn.k_proj,0.0000022945,0.01000,1.484
|
|
44,self_attn.v_proj,0.0000024551,0.01000,1.512
|
|
44,self_attn.q_proj,0.0000470545,0.01000,1.537
|
|
44,self_attn.o_proj,0.0000083398,0.01000,1.520
|
|
44,mlp.gate_up_proj,0.0000717656,0.01000,1.758
|
|
44,mlp.down_proj,0.0000031455,0.01000,7.725
|
|
45,self_attn.k_proj,0.0000026021,0.01000,1.494
|
|
45,self_attn.v_proj,0.0000024987,0.01000,1.522
|
|
45,self_attn.q_proj,0.0000503206,0.01000,1.538
|
|
45,self_attn.o_proj,0.0000093540,0.01000,1.514
|
|
45,mlp.gate_up_proj,0.0000653584,0.01000,1.777
|
|
45,mlp.down_proj,0.0000027631,0.01000,7.734
|
|
46,self_attn.k_proj,0.0000023923,0.01000,1.496
|
|
46,self_attn.v_proj,0.0000025303,0.01000,1.519
|
|
46,self_attn.q_proj,0.0000464908,0.01000,1.534
|
|
46,self_attn.o_proj,0.0000092665,0.01000,1.522
|
|
46,mlp.gate_up_proj,0.0000673613,0.01000,1.763
|
|
46,mlp.down_proj,0.0000032540,0.01000,7.743
|
|
47,self_attn.k_proj,0.0000022541,0.01000,1.482
|
|
47,self_attn.v_proj,0.0000024099,0.01000,1.507
|
|
47,self_attn.q_proj,0.0000424197,0.01000,1.529
|
|
47,self_attn.o_proj,0.0000087173,0.01000,1.520
|
|
47,mlp.gate_up_proj,0.0000696124,0.01000,1.758
|
|
47,mlp.down_proj,0.0000032250,0.01000,7.752
|
|
48,self_attn.k_proj,0.0000022253,0.01000,1.492
|
|
48,self_attn.v_proj,0.0000022728,0.01000,1.514
|
|
48,self_attn.q_proj,0.0000427026,0.01000,1.553
|
|
48,self_attn.o_proj,0.0000073344,0.01000,1.540
|
|
48,mlp.gate_up_proj,0.0000666436,0.01000,1.771
|
|
48,mlp.down_proj,0.0000034516,0.01000,7.751
|
|
49,self_attn.k_proj,0.0000023598,0.01000,1.489
|
|
49,self_attn.v_proj,0.0000025774,0.01000,1.522
|
|
49,self_attn.q_proj,0.0000439942,0.01000,1.545
|
|
49,self_attn.o_proj,0.0000058884,0.01000,1.524
|
|
49,mlp.gate_up_proj,0.0001054278,0.01000,1.762
|
|
49,mlp.down_proj,0.0000066771,0.01000,7.784
|
|
50,self_attn.k_proj,0.0000023662,0.01000,1.478
|
|
50,self_attn.v_proj,0.0000027904,0.01000,1.504
|
|
50,self_attn.q_proj,0.0000424220,0.01000,1.526
|
|
50,self_attn.o_proj,0.0000072030,0.01000,1.516
|
|
50,mlp.gate_up_proj,0.0001290033,0.01000,1.755
|
|
50,mlp.down_proj,0.0000091810,0.01000,7.675
|
|
51,self_attn.k_proj,0.0000022649,0.01000,1.489
|
|
51,self_attn.v_proj,0.0000022110,0.01000,1.517
|
|
51,self_attn.q_proj,0.0000381388,0.01000,1.558
|
|
51,self_attn.o_proj,0.0000036714,0.01000,1.539
|
|
51,mlp.gate_up_proj,0.0001260655,0.01000,1.749
|
|
51,mlp.down_proj,0.0000091918,0.01000,7.712
|
|
52,self_attn.k_proj,0.0000024394,0.01000,1.474
|
|
52,self_attn.v_proj,0.0000023174,0.01000,1.500
|
|
52,self_attn.q_proj,0.0000422424,0.01000,1.547
|
|
52,self_attn.o_proj,0.0000042098,0.01000,1.526
|
|
52,mlp.gate_up_proj,0.0001209586,0.01000,1.776
|
|
52,mlp.down_proj,0.0000086012,0.01000,7.724
|
|
53,self_attn.k_proj,0.0000020518,0.01000,1.503
|
|
53,self_attn.v_proj,0.0000024131,0.01000,1.522
|
|
53,self_attn.q_proj,0.0000456590,0.01000,1.566
|
|
53,self_attn.o_proj,0.0000017261,0.01000,1.529
|
|
53,mlp.gate_up_proj,0.0001240569,0.01000,1.771
|
|
53,mlp.down_proj,0.0000086225,0.01000,7.725
|
|
54,self_attn.k_proj,0.0000021914,0.01000,1.496
|
|
54,self_attn.v_proj,0.0000026488,0.01000,1.550
|
|
54,self_attn.q_proj,0.0000383103,0.01000,1.556
|
|
54,self_attn.o_proj,0.0000011243,0.01000,1.553
|
|
54,mlp.gate_up_proj,0.0001322332,0.01000,1.801
|
|
54,mlp.down_proj,0.0000096043,0.01000,7.784
|
|
55,self_attn.k_proj,0.0000021482,0.01000,1.485
|
|
55,self_attn.v_proj,0.0000023087,0.01000,1.506
|
|
55,self_attn.q_proj,0.0000335327,0.01000,1.527
|
|
55,self_attn.o_proj,0.0000010707,0.01000,1.513
|
|
55,mlp.gate_up_proj,0.0001262408,0.01000,1.763
|
|
55,mlp.down_proj,0.0000091236,0.01000,7.742
|
|
56,self_attn.k_proj,0.0000020794,0.01000,1.475
|
|
56,self_attn.v_proj,0.0000026723,0.01000,1.510
|
|
56,self_attn.q_proj,0.0000372959,0.01000,1.531
|
|
56,self_attn.o_proj,0.0000023217,0.01000,1.514
|
|
56,mlp.gate_up_proj,0.0001405949,0.01000,1.746
|
|
56,mlp.down_proj,0.0000117001,0.01000,7.710
|
|
57,self_attn.k_proj,0.0000021717,0.01000,1.477
|
|
57,self_attn.v_proj,0.0000020618,0.01000,1.543
|
|
57,self_attn.q_proj,0.0000394623,0.01000,1.537
|
|
57,self_attn.o_proj,0.0000070200,0.01000,1.507
|
|
57,mlp.gate_up_proj,0.0001650417,0.01000,1.744
|
|
57,mlp.down_proj,0.0000167601,0.01000,7.679
|
|
58,self_attn.k_proj,0.0000021242,0.01000,1.501
|
|
58,self_attn.v_proj,0.0000021589,0.01000,1.528
|
|
58,self_attn.q_proj,0.0000392971,0.01000,1.574
|
|
58,self_attn.o_proj,0.0000022084,0.01000,1.542
|
|
58,mlp.gate_up_proj,0.0002288546,0.01000,1.770
|
|
58,mlp.down_proj,0.0000346537,0.01000,7.762
|
|
59,self_attn.k_proj,0.0000021596,0.01000,1.490
|
|
59,self_attn.v_proj,0.0000019384,0.01000,1.514
|
|
59,self_attn.q_proj,0.0000360824,0.01000,1.533
|
|
59,self_attn.o_proj,0.0000036960,0.01000,1.520
|
|
59,mlp.gate_up_proj,0.0002469802,0.01000,1.763
|
|
59,mlp.down_proj,0.0000465371,0.01000,7.804
|
|
60,self_attn.k_proj,0.0000019967,0.01000,1.491
|
|
60,self_attn.v_proj,0.0000018081,0.01000,1.523
|
|
60,self_attn.q_proj,0.0000359417,0.01000,1.548
|
|
60,self_attn.o_proj,0.0000057281,0.01000,1.521
|
|
60,mlp.gate_up_proj,0.0002825791,0.01000,1.766
|
|
60,mlp.down_proj,0.0000614542,0.01000,7.746
|
|
|