Abdohaaland's picture
Upload folder using huggingface_hub
ee77be3 verified
layer,module,loss,samples,damp,time
0,self_attn.qkv_proj,0.0000033241,0.05000,1.466
0,self_attn.o_proj,0.0000001653,0.05000,0.972
0,mlp.up_proj,0.0000088873,0.05000,2.326
0,mlp.gate_proj,0.0000093133,0.05000,2.381
0,mlp.down_proj,0.0000006132,0.05000,2.897
1,self_attn.qkv_proj,0.0000256600,0.05000,1.133
1,self_attn.o_proj,0.0000002638,0.05000,1.190
1,mlp.up_proj,0.0000117479,0.05000,2.989
1,mlp.gate_proj,0.0000127222,0.05000,3.056
1,mlp.down_proj,0.0000011313,0.05000,2.995
2,self_attn.qkv_proj,0.0000549695,0.05000,1.098
2,self_attn.o_proj,0.0000003222,0.05000,1.060
2,mlp.up_proj,0.0000164835,0.05000,2.991
2,mlp.gate_proj,0.0000180576,0.05000,3.067
2,mlp.down_proj,0.0000009932,0.05000,2.948
3,self_attn.qkv_proj,0.0000797589,0.05000,1.024
3,self_attn.o_proj,0.0000005779,0.05000,1.035
3,mlp.gate_proj,0.0000227424,0.05000,2.564
3,mlp.up_proj,0.0000200395,0.05000,2.588
3,mlp.down_proj,0.0002782354,0.05000,2.898
4,self_attn.qkv_proj,0.0000732176,0.05000,1.017
4,self_attn.o_proj,0.0000009411,0.05000,1.014
4,mlp.gate_proj,0.0000293015,0.05000,2.410
4,mlp.up_proj,0.0000244080,0.05000,2.435
4,mlp.down_proj,0.0000023525,0.05000,2.982
5,self_attn.qkv_proj,0.0000933612,0.05000,1.026
5,self_attn.o_proj,0.0000017404,0.05000,1.043
5,mlp.gate_proj,0.0000357003,0.05000,2.498
5,mlp.up_proj,0.0000289614,0.05000,2.512
5,mlp.down_proj,0.0000036120,0.05000,2.905
6,self_attn.qkv_proj,0.0000769730,0.05000,1.034
6,self_attn.o_proj,0.0000019431,0.05000,0.999
6,mlp.gate_proj,0.0000366409,0.05000,2.308
6,mlp.up_proj,0.0000313463,0.05000,2.333
6,mlp.down_proj,0.0000043249,0.05000,2.952
7,self_attn.qkv_proj,0.0001243836,0.05000,1.054
7,self_attn.o_proj,0.0000031223,0.05000,1.024
7,mlp.up_proj,0.0000345948,0.05000,2.341
7,mlp.gate_proj,0.0000385413,0.05000,2.381
7,mlp.down_proj,0.0000056692,0.05000,2.957
8,self_attn.qkv_proj,0.0001220055,0.05000,1.034
8,self_attn.o_proj,0.0000041093,0.05000,1.017
8,mlp.up_proj,0.0000381113,0.05000,2.440
8,mlp.gate_proj,0.0000393560,0.05000,2.468
8,mlp.down_proj,0.0000063462,0.05000,2.941
9,self_attn.qkv_proj,0.0001055136,0.05000,1.018
9,self_attn.o_proj,0.0000038115,0.05000,1.023
9,mlp.up_proj,0.0000413347,0.05000,2.383
9,mlp.gate_proj,0.0000413403,0.05000,2.401
9,mlp.down_proj,0.0000060606,0.05000,2.969
10,self_attn.qkv_proj,0.0001138408,0.05000,1.028
10,self_attn.o_proj,0.0000050139,0.05000,1.026
10,mlp.up_proj,0.0000424914,0.05000,2.415
10,mlp.gate_proj,0.0000427714,0.05000,2.426
10,mlp.down_proj,0.0000062799,0.05000,3.024
11,self_attn.qkv_proj,0.0001136194,0.05000,1.025
11,self_attn.o_proj,0.0000053294,0.05000,1.033
11,mlp.gate_proj,0.0000435245,0.05000,2.301
11,mlp.up_proj,0.0000456607,0.05000,2.340
11,mlp.down_proj,0.0000071477,0.05000,2.955
12,self_attn.qkv_proj,0.0001161936,0.05000,1.074
12,self_attn.o_proj,0.0000050933,0.05000,1.018
12,mlp.gate_proj,0.0000425567,0.05000,2.328
12,mlp.up_proj,0.0000467951,0.05000,2.354
12,mlp.down_proj,0.0000079555,0.05000,2.941
13,self_attn.qkv_proj,0.0001463183,0.05000,1.023
13,self_attn.o_proj,0.0000080918,0.05000,1.013
13,mlp.up_proj,0.0000516755,0.05000,2.454
13,mlp.gate_proj,0.0000449235,0.05000,2.467
13,mlp.down_proj,0.0000089921,0.05000,2.883
14,self_attn.qkv_proj,0.0001305041,0.05000,1.017
14,self_attn.o_proj,0.0000074252,0.05000,1.019
14,mlp.gate_proj,0.0000498036,0.05000,2.374
14,mlp.up_proj,0.0000568754,0.05000,2.399
14,mlp.down_proj,0.0000102694,0.05000,2.953
15,self_attn.qkv_proj,0.0001450314,0.05000,1.016
15,self_attn.o_proj,0.0000070636,0.05000,1.018
15,mlp.gate_proj,0.0000541943,0.05000,2.286
15,mlp.up_proj,0.0000617730,0.05000,2.306
15,mlp.down_proj,0.0000115088,0.05000,2.977
16,self_attn.qkv_proj,0.0001367455,0.05000,1.021
16,self_attn.o_proj,0.0000106811,0.05000,1.077
16,mlp.gate_proj,0.0000558458,0.05000,2.282
16,mlp.up_proj,0.0000641705,0.05000,2.314
16,mlp.down_proj,0.0000146563,0.05000,2.944
17,self_attn.qkv_proj,0.0001290727,0.05000,1.030
17,self_attn.o_proj,0.0000075975,0.05000,1.049
17,mlp.gate_proj,0.0000583090,0.05000,2.453
17,mlp.up_proj,0.0000678901,0.05000,2.486
17,mlp.down_proj,0.0000175746,0.05000,2.974
18,self_attn.qkv_proj,0.0001274209,0.05000,1.018
18,self_attn.o_proj,0.0000124687,0.05000,1.052
18,mlp.up_proj,0.0000756719,0.05000,2.402
18,mlp.gate_proj,0.0000658952,0.05000,2.437
18,mlp.down_proj,0.0000258305,0.05000,2.968
19,self_attn.qkv_proj,0.0001435843,0.05000,1.026
19,self_attn.o_proj,0.0000158486,0.05000,1.022
19,mlp.up_proj,0.0000815260,0.05000,2.438
19,mlp.gate_proj,0.0000720414,0.05000,2.465
19,mlp.down_proj,0.0000294331,0.05000,2.928
20,self_attn.qkv_proj,0.0001412000,0.05000,1.041
20,self_attn.o_proj,0.0000198388,0.05000,1.020
20,mlp.gate_proj,0.0000866334,0.05000,2.407
20,mlp.up_proj,0.0000980211,0.05000,2.433
20,mlp.down_proj,0.0000357816,0.05000,2.969
21,self_attn.qkv_proj,0.0001619060,0.05000,1.030
21,self_attn.o_proj,0.0000168971,0.05000,1.010
21,mlp.gate_proj,0.0000912907,0.05000,2.397
21,mlp.up_proj,0.0001050032,0.05000,2.420
21,mlp.down_proj,0.0000443997,0.05000,2.936
22,self_attn.qkv_proj,0.0001537729,0.05000,1.016
22,self_attn.o_proj,0.0000249352,0.05000,1.006
22,mlp.up_proj,0.0001121257,0.05000,2.376
22,mlp.gate_proj,0.0000962384,0.05000,2.379
22,mlp.down_proj,0.0000494136,0.05000,2.923
23,self_attn.qkv_proj,0.0001589419,0.05000,1.051
23,self_attn.o_proj,0.0000104669,0.05000,1.009
23,mlp.gate_proj,0.0000999043,0.05000,2.438
23,mlp.up_proj,0.0001153954,0.05000,2.463
23,mlp.down_proj,0.0000453592,0.05000,2.927
24,self_attn.qkv_proj,0.0001645199,0.05000,1.031
24,self_attn.o_proj,0.0000096019,0.05000,1.021
24,mlp.up_proj,0.0001215475,0.05000,2.423
24,mlp.gate_proj,0.0001040655,0.05000,2.457
24,mlp.down_proj,0.0000538232,0.05000,2.972
25,self_attn.qkv_proj,0.0001800437,0.05000,1.046
25,self_attn.o_proj,0.0000219898,0.05000,1.034
25,mlp.up_proj,0.0001317978,0.05000,2.360
25,mlp.gate_proj,0.0001123926,0.05000,2.376
25,mlp.down_proj,0.0000615177,0.05000,3.004
26,self_attn.qkv_proj,0.0001578350,0.05000,1.045
26,self_attn.o_proj,0.0000382162,0.05000,1.019
26,mlp.up_proj,0.0001344800,0.05000,2.456
26,mlp.gate_proj,0.0001149784,0.05000,2.460
26,mlp.down_proj,0.0000757419,0.05000,2.945
27,self_attn.qkv_proj,0.0001475294,0.05000,1.035
27,self_attn.o_proj,0.0000206933,0.05000,1.017
27,mlp.up_proj,0.0001520072,0.05000,2.336
27,mlp.gate_proj,0.0001321883,0.05000,2.361
27,mlp.down_proj,0.0001098592,0.05000,2.962
28,self_attn.qkv_proj,0.0002010034,0.05000,1.035
28,self_attn.o_proj,0.0000312271,0.05000,1.009
28,mlp.gate_proj,0.0001526327,0.05000,2.282
28,mlp.up_proj,0.0001689870,0.05000,2.303
28,mlp.down_proj,0.0001587952,0.05000,2.939
29,self_attn.qkv_proj,0.0001802269,0.05000,1.036
29,self_attn.o_proj,0.0000341774,0.05000,1.013
29,mlp.gate_proj,0.0001712056,0.05000,2.437
29,mlp.up_proj,0.0001709735,0.05000,2.456
29,mlp.down_proj,0.0002735091,0.05000,2.917
30,self_attn.qkv_proj,0.0001263867,0.05000,1.014
30,self_attn.o_proj,0.0000344519,0.05000,1.018
30,mlp.up_proj,0.0001645436,0.05000,2.287
30,mlp.gate_proj,0.0001920412,0.05000,2.311
30,mlp.down_proj,0.0014165752,0.05000,2.951
31,self_attn.qkv_proj,0.0000026388,0.05000,1.010
31,self_attn.o_proj,0.0000037210,0.05000,1.014
31,mlp.up_proj,0.0000405552,0.05000,2.355
31,mlp.gate_proj,0.0000741658,0.05000,2.395
31,mlp.down_proj,0.0004612778,0.05000,2.959