Abdohaaland's picture
Upload folder using huggingface_hub
53bb6fd verified
layer,module,loss,samples,damp,time
0,self_attn.qkv_proj,0.0000029929,0.05000,1.589
0,self_attn.o_proj,0.0000001045,0.05000,0.964
0,mlp.gate_proj,0.0000063314,0.05000,2.138
0,mlp.up_proj,0.0000059726,0.05000,2.171
0,mlp.down_proj,0.0000001969,0.05000,2.824
1,self_attn.qkv_proj,0.0000177958,0.05000,1.157
1,self_attn.o_proj,0.0000001157,0.05000,1.036
1,mlp.gate_proj,0.0000088358,0.05000,2.975
1,mlp.up_proj,0.0000080169,0.05000,3.019
1,mlp.down_proj,0.0000003500,0.05000,2.972
2,self_attn.qkv_proj,0.0000283670,0.05000,1.035
2,self_attn.o_proj,0.0000001541,0.05000,1.075
2,mlp.gate_proj,0.0000087870,0.05000,2.253
2,mlp.up_proj,0.0000078665,0.05000,2.265
2,mlp.down_proj,0.0000003255,0.05000,2.942
3,self_attn.qkv_proj,0.0000338127,0.05000,1.005
3,self_attn.o_proj,0.0000003137,0.05000,1.011
3,mlp.gate_proj,0.0000158869,0.05000,2.163
3,mlp.up_proj,0.0000136104,0.05000,2.196
3,mlp.down_proj,0.0000427317,0.05000,2.914
4,self_attn.qkv_proj,0.0000373079,0.05000,1.048
4,self_attn.o_proj,0.0000003564,0.05000,0.996
4,mlp.up_proj,0.0000131388,0.05000,2.286
4,mlp.gate_proj,0.0000163559,0.05000,2.303
4,mlp.down_proj,0.0000009839,0.05000,2.888
5,self_attn.qkv_proj,0.0000445573,0.05000,1.013
5,self_attn.o_proj,0.0000005962,0.05000,1.023
5,mlp.gate_proj,0.0000180498,0.05000,2.378
5,mlp.up_proj,0.0000140669,0.05000,2.414
5,mlp.down_proj,0.0000012192,0.05000,2.923
6,self_attn.qkv_proj,0.0000341724,0.05000,1.010
6,self_attn.o_proj,0.0000011619,0.05000,1.020
6,mlp.gate_proj,0.0000183702,0.05000,2.365
6,mlp.up_proj,0.0000152781,0.05000,2.367
6,mlp.down_proj,0.0000016868,0.05000,2.890
7,self_attn.qkv_proj,0.0000553964,0.05000,1.012
7,self_attn.o_proj,0.0000007415,0.05000,0.993
7,mlp.gate_proj,0.0000174044,0.05000,2.255
7,mlp.up_proj,0.0000152908,0.05000,2.276
7,mlp.down_proj,0.0000016985,0.05000,2.981
8,self_attn.qkv_proj,0.0000495149,0.05000,1.002
8,self_attn.o_proj,0.0000006034,0.05000,0.999
8,mlp.gate_proj,0.0000178227,0.05000,2.266
8,mlp.up_proj,0.0000171840,0.05000,2.289
8,mlp.down_proj,0.0000018485,0.05000,2.960
9,self_attn.qkv_proj,0.0000430816,0.05000,1.006
9,self_attn.o_proj,0.0000009036,0.05000,1.081
9,mlp.gate_proj,0.0000173485,0.05000,2.280
9,mlp.up_proj,0.0000173551,0.05000,2.298
9,mlp.down_proj,0.0000018817,0.05000,2.903
10,self_attn.qkv_proj,0.0000433983,0.05000,1.010
10,self_attn.o_proj,0.0000012253,0.05000,1.009
10,mlp.up_proj,0.0000191325,0.05000,2.302
10,mlp.gate_proj,0.0000193263,0.05000,2.319
10,mlp.down_proj,0.0000021741,0.05000,2.882
11,self_attn.qkv_proj,0.0000463978,0.05000,1.046
11,self_attn.o_proj,0.0000018500,0.05000,0.981
11,mlp.up_proj,0.0000213888,0.05000,2.286
11,mlp.gate_proj,0.0000202338,0.05000,2.311
11,mlp.down_proj,0.0000025080,0.05000,2.888
12,self_attn.qkv_proj,0.0000507519,0.05000,1.118
12,self_attn.o_proj,0.0000016432,0.05000,1.053
12,mlp.gate_proj,0.0000210578,0.05000,2.404
12,mlp.up_proj,0.0000233956,0.05000,2.436
12,mlp.down_proj,0.0000030367,0.05000,3.036
13,self_attn.qkv_proj,0.0000660197,0.05000,1.024
13,self_attn.o_proj,0.0000018919,0.05000,1.027
13,mlp.gate_proj,0.0000217064,0.05000,2.295
13,mlp.up_proj,0.0000255225,0.05000,2.312
13,mlp.down_proj,0.0000033143,0.05000,2.927
14,self_attn.qkv_proj,0.0000574443,0.05000,0.995
14,self_attn.o_proj,0.0000015500,0.05000,1.016
14,mlp.up_proj,0.0000290123,0.05000,2.244
14,mlp.gate_proj,0.0000248531,0.05000,2.264
14,mlp.down_proj,0.0000037807,0.05000,2.944
15,self_attn.qkv_proj,0.0000651020,0.05000,1.009
15,self_attn.o_proj,0.0000026469,0.05000,1.033
15,mlp.up_proj,0.0000307608,0.05000,2.212
15,mlp.gate_proj,0.0000263729,0.05000,2.240
15,mlp.down_proj,0.0000042095,0.05000,2.912
16,self_attn.qkv_proj,0.0000577900,0.05000,1.016
16,self_attn.o_proj,0.0000031993,0.05000,1.026
16,mlp.up_proj,0.0000332469,0.05000,2.312
16,mlp.gate_proj,0.0000283605,0.05000,2.331
16,mlp.down_proj,0.0000052141,0.05000,2.921
17,self_attn.qkv_proj,0.0000622398,0.05000,1.017
17,self_attn.o_proj,0.0000029537,0.05000,1.056
17,mlp.gate_proj,0.0000317461,0.05000,2.302
17,mlp.up_proj,0.0000378686,0.05000,2.327
17,mlp.down_proj,0.0000068786,0.05000,2.916
18,self_attn.qkv_proj,0.0000657631,0.05000,1.011
18,self_attn.o_proj,0.0000065101,0.05000,1.037
18,mlp.up_proj,0.0000418297,0.05000,2.208
18,mlp.gate_proj,0.0000357805,0.05000,2.237
18,mlp.down_proj,0.0000089543,0.05000,2.894
19,self_attn.qkv_proj,0.0000770983,0.05000,1.020
19,self_attn.o_proj,0.0000077340,0.05000,1.028
19,mlp.up_proj,0.0000444914,0.05000,2.382
19,mlp.gate_proj,0.0000386684,0.05000,2.380
19,mlp.down_proj,0.0000109362,0.05000,2.911
20,self_attn.qkv_proj,0.0000740929,0.05000,1.003
20,self_attn.o_proj,0.0000081113,0.05000,0.986
20,mlp.gate_proj,0.0000472041,0.05000,2.307
20,mlp.up_proj,0.0000543119,0.05000,2.330
20,mlp.down_proj,0.0000145892,0.05000,2.990
21,self_attn.qkv_proj,0.0000857731,0.05000,1.021
21,self_attn.o_proj,0.0000068950,0.05000,0.998
21,mlp.up_proj,0.0000586656,0.05000,2.283
21,mlp.gate_proj,0.0000500638,0.05000,2.313
21,mlp.down_proj,0.0000176288,0.05000,2.957
22,self_attn.qkv_proj,0.0000867742,0.05000,1.019
22,self_attn.o_proj,0.0000102738,0.05000,0.996
22,mlp.up_proj,0.0000663261,0.05000,2.281
22,mlp.gate_proj,0.0000555567,0.05000,2.303
22,mlp.down_proj,0.0000202035,0.05000,2.910
23,self_attn.qkv_proj,0.0000962782,0.05000,1.005
23,self_attn.o_proj,0.0000096911,0.05000,1.009
23,mlp.up_proj,0.0000733566,0.05000,2.193
23,mlp.gate_proj,0.0000619966,0.05000,2.214
23,mlp.down_proj,0.0000254390,0.05000,2.972
24,self_attn.qkv_proj,0.0001047326,0.05000,1.043
24,self_attn.o_proj,0.0000094759,0.05000,0.995
24,mlp.up_proj,0.0000819970,0.05000,2.187
24,mlp.gate_proj,0.0000681171,0.05000,2.217
24,mlp.down_proj,0.0000320756,0.05000,2.875
25,self_attn.qkv_proj,0.0001196687,0.05000,1.032
25,self_attn.o_proj,0.0000178841,0.05000,0.992
25,mlp.gate_proj,0.0000786205,0.05000,2.282
25,mlp.up_proj,0.0000949818,0.05000,2.302
25,mlp.down_proj,0.0000417770,0.05000,2.970
26,self_attn.qkv_proj,0.0001117620,0.05000,1.023
26,self_attn.o_proj,0.0000252745,0.05000,0.996
26,mlp.up_proj,0.0001003381,0.05000,2.225
26,mlp.gate_proj,0.0000833383,0.05000,2.252
26,mlp.down_proj,0.0000515722,0.05000,2.924
27,self_attn.qkv_proj,0.0001086565,0.05000,1.016
27,self_attn.o_proj,0.0000264664,0.05000,0.999
27,mlp.up_proj,0.0001091925,0.05000,2.223
27,mlp.gate_proj,0.0000928646,0.05000,2.222
27,mlp.down_proj,0.0000621013,0.05000,2.919
28,self_attn.qkv_proj,0.0001426622,0.05000,1.013
28,self_attn.o_proj,0.0000346105,0.05000,0.997
28,mlp.up_proj,0.0001217435,0.05000,2.149
28,mlp.gate_proj,0.0001090677,0.05000,2.162
28,mlp.down_proj,0.0000804849,0.05000,2.925
29,self_attn.qkv_proj,0.0001282340,0.05000,1.021
29,self_attn.o_proj,0.0000445930,0.05000,1.029
29,mlp.up_proj,0.0001171631,0.05000,2.218
29,mlp.gate_proj,0.0001193069,0.05000,2.254
29,mlp.down_proj,0.0000950395,0.05000,2.887
30,self_attn.qkv_proj,0.0001681025,0.05000,0.995
30,self_attn.o_proj,0.0000827839,0.05000,1.008
30,mlp.up_proj,0.0002124111,0.05000,2.299
30,mlp.gate_proj,0.0002344202,0.05000,2.309
30,mlp.down_proj,1.6844320378,0.05000,2.899
31,self_attn.qkv_proj,0.0000036079,0.05000,1.055
31,self_attn.o_proj,0.0000053948,0.05000,1.004
31,mlp.gate_proj,0.0000985641,0.05000,2.237
31,mlp.up_proj,0.0000607404,0.05000,2.259
31,mlp.down_proj,0.0000745720,0.05000,2.906