Abdohaaland's picture
Upload folder using huggingface_hub
6a102df verified
layer,module,loss,samples,damp,time
0,self_attn.qkv_proj,0.0000033337,0.05000,1.970
0,self_attn.o_proj,0.0000001568,0.05000,1.193
0,mlp.gate_proj,0.0000094541,0.05000,2.477
0,mlp.up_proj,0.0000090174,0.05000,2.527
0,mlp.down_proj,0.0000005734,0.05000,2.784
1,self_attn.qkv_proj,0.0000253673,0.05000,1.153
1,self_attn.o_proj,0.0000003501,0.05000,1.191
1,mlp.up_proj,0.0000115117,0.05000,2.958
1,mlp.gate_proj,0.0000125914,0.05000,2.991
1,mlp.down_proj,0.0000009363,0.05000,3.015
2,self_attn.qkv_proj,0.0000545575,0.05000,1.052
2,self_attn.o_proj,0.0000003618,0.05000,1.068
2,mlp.up_proj,0.0000162658,0.05000,3.042
2,mlp.gate_proj,0.0000179054,0.05000,3.061
2,mlp.down_proj,0.0000010033,0.05000,2.914
3,self_attn.qkv_proj,0.0000754067,0.05000,1.075
3,self_attn.o_proj,0.0000005250,0.05000,0.998
3,mlp.up_proj,0.0000187377,0.05000,2.269
3,mlp.gate_proj,0.0000214041,0.05000,2.297
3,mlp.down_proj,0.0002351547,0.05000,2.890
4,self_attn.qkv_proj,0.0000709153,0.05000,1.011
4,self_attn.o_proj,0.0000009426,0.05000,1.000
4,mlp.up_proj,0.0000266585,0.05000,2.434
4,mlp.gate_proj,0.0000322079,0.05000,2.466
4,mlp.down_proj,0.0000028561,0.05000,2.926
5,self_attn.qkv_proj,0.0000920992,0.05000,1.010
5,self_attn.o_proj,0.0000015057,0.05000,1.010
5,mlp.up_proj,0.0000311401,0.05000,2.282
5,mlp.gate_proj,0.0000387061,0.05000,2.308
5,mlp.down_proj,0.0000040372,0.05000,2.975
6,self_attn.qkv_proj,0.0000751325,0.05000,1.011
6,self_attn.o_proj,0.0000017975,0.05000,1.018
6,mlp.gate_proj,0.0000391059,0.05000,2.604
6,mlp.up_proj,0.0000332029,0.05000,2.646
6,mlp.down_proj,0.0000049817,0.05000,2.970
7,self_attn.qkv_proj,0.0001157685,0.05000,1.043
7,self_attn.o_proj,0.0000020251,0.05000,1.062
7,mlp.up_proj,0.0000352015,0.05000,2.432
7,mlp.gate_proj,0.0000395090,0.05000,2.451
7,mlp.down_proj,0.0000058283,0.05000,2.930
8,self_attn.qkv_proj,0.0001045483,0.05000,1.025
8,self_attn.o_proj,0.0000020512,0.05000,1.005
8,mlp.gate_proj,0.0000368261,0.05000,2.249
8,mlp.up_proj,0.0000355081,0.05000,2.276
8,mlp.down_proj,0.0000057011,0.05000,2.933
9,self_attn.qkv_proj,0.0000903360,0.05000,1.021
9,self_attn.o_proj,0.0000025147,0.05000,1.024
9,mlp.gate_proj,0.0000381440,0.05000,2.258
9,mlp.up_proj,0.0000379692,0.05000,2.287
9,mlp.down_proj,0.0000054069,0.05000,2.922
10,self_attn.qkv_proj,0.0000977697,0.05000,1.010
10,self_attn.o_proj,0.0000037061,0.05000,1.050
10,mlp.up_proj,0.0000417211,0.05000,2.357
10,mlp.gate_proj,0.0000421710,0.05000,2.380
10,mlp.down_proj,0.0000064649,0.05000,2.948
11,self_attn.qkv_proj,0.0001041781,0.05000,1.010
11,self_attn.o_proj,0.0000043695,0.05000,1.035
11,mlp.gate_proj,0.0000435187,0.05000,2.366
11,mlp.up_proj,0.0000454773,0.05000,2.381
11,mlp.down_proj,0.0000071585,0.05000,2.931
12,self_attn.qkv_proj,0.0001108157,0.05000,1.013
12,self_attn.o_proj,0.0000038909,0.05000,1.009
12,mlp.gate_proj,0.0000440640,0.05000,2.264
12,mlp.up_proj,0.0000483411,0.05000,2.282
12,mlp.down_proj,0.0000083439,0.05000,2.919
13,self_attn.qkv_proj,0.0001360751,0.05000,1.019
13,self_attn.o_proj,0.0000052360,0.05000,0.993
13,mlp.gate_proj,0.0000448895,0.05000,2.295
13,mlp.up_proj,0.0000516026,0.05000,2.308
13,mlp.down_proj,0.0000088433,0.05000,2.981
14,self_attn.qkv_proj,0.0001205465,0.05000,1.058
14,self_attn.o_proj,0.0000045024,0.05000,1.012
14,mlp.gate_proj,0.0000490126,0.05000,2.358
14,mlp.up_proj,0.0000558636,0.05000,2.385
14,mlp.down_proj,0.0000094944,0.05000,2.907
15,self_attn.qkv_proj,0.0001328261,0.05000,1.018
15,self_attn.o_proj,0.0000043096,0.05000,1.019
15,mlp.gate_proj,0.0000536100,0.05000,2.319
15,mlp.up_proj,0.0000610305,0.05000,2.339
15,mlp.down_proj,0.0000106724,0.05000,2.888
16,self_attn.qkv_proj,0.0001251596,0.05000,1.025
16,self_attn.o_proj,0.0000077229,0.05000,1.004
16,mlp.up_proj,0.0000627541,0.05000,2.347
16,mlp.gate_proj,0.0000547321,0.05000,2.372
16,mlp.down_proj,0.0000136088,0.05000,2.974
17,self_attn.qkv_proj,0.0001212124,0.05000,1.015
17,self_attn.o_proj,0.0000051576,0.05000,1.019
17,mlp.gate_proj,0.0000581634,0.05000,2.241
17,mlp.up_proj,0.0000677224,0.05000,2.266
17,mlp.down_proj,0.0000170442,0.05000,2.977
18,self_attn.qkv_proj,0.0001200134,0.05000,0.988
18,self_attn.o_proj,0.0000090346,0.05000,1.000
18,mlp.gate_proj,0.0000653781,0.05000,2.245
18,mlp.up_proj,0.0000751556,0.05000,2.282
18,mlp.down_proj,0.0000247040,0.05000,2.888
19,self_attn.qkv_proj,0.0001405187,0.05000,1.018
19,self_attn.o_proj,0.0000128052,0.05000,1.005
19,mlp.up_proj,0.0000819399,0.05000,2.238
19,mlp.gate_proj,0.0000723034,0.05000,2.255
19,mlp.down_proj,0.0000287014,0.05000,2.892
20,self_attn.qkv_proj,0.0001370618,0.05000,1.049
20,self_attn.o_proj,0.0000141441,0.05000,0.995
20,mlp.gate_proj,0.0000871970,0.05000,2.277
20,mlp.up_proj,0.0000987938,0.05000,2.300
20,mlp.down_proj,0.0000353455,0.05000,2.886
21,self_attn.qkv_proj,0.0001576805,0.05000,1.020
21,self_attn.o_proj,0.0000103696,0.05000,0.997
21,mlp.gate_proj,0.0000898126,0.05000,2.334
21,mlp.up_proj,0.0001034518,0.05000,2.360
21,mlp.down_proj,0.0000406898,0.05000,2.901
22,self_attn.qkv_proj,0.0001496053,0.05000,1.007
22,self_attn.o_proj,0.0000179317,0.05000,1.007
22,mlp.up_proj,0.0001094033,0.05000,2.271
22,mlp.gate_proj,0.0000937537,0.05000,2.296
22,mlp.down_proj,0.0000450660,0.05000,2.890
23,self_attn.qkv_proj,0.0001562641,0.05000,1.000
23,self_attn.o_proj,0.0000076899,0.05000,1.000
23,mlp.up_proj,0.0001155189,0.05000,2.287
23,mlp.gate_proj,0.0000998504,0.05000,2.312
23,mlp.down_proj,0.0000445921,0.05000,2.932
24,self_attn.qkv_proj,0.0001635886,0.05000,1.013
24,self_attn.o_proj,0.0000070508,0.05000,1.015
24,mlp.up_proj,0.0001205805,0.05000,2.324
24,mlp.gate_proj,0.0001030940,0.05000,2.348
24,mlp.down_proj,0.0000522420,0.05000,2.951
25,self_attn.qkv_proj,0.0001784758,0.05000,1.019
25,self_attn.o_proj,0.0000145658,0.05000,0.995
25,mlp.up_proj,0.0001376533,0.05000,2.229
25,mlp.gate_proj,0.0001171439,0.05000,2.260
25,mlp.down_proj,0.0000661028,0.05000,2.879
26,self_attn.qkv_proj,0.0001639939,0.05000,1.066
26,self_attn.o_proj,0.0000326306,0.05000,1.007
26,mlp.gate_proj,0.0001198269,0.05000,2.302
26,mlp.up_proj,0.0001404009,0.05000,2.325
26,mlp.down_proj,0.0000833422,0.05000,2.896
27,self_attn.qkv_proj,0.0001548430,0.05000,1.001
27,self_attn.o_proj,0.0000179886,0.05000,1.018
27,mlp.gate_proj,0.0001427162,0.05000,2.394
27,mlp.up_proj,0.0001644143,0.05000,2.412
27,mlp.down_proj,0.0001287594,0.05000,2.911
28,self_attn.qkv_proj,0.0002212718,0.05000,1.007
28,self_attn.o_proj,0.0000247850,0.05000,1.008
28,mlp.up_proj,0.0001861897,0.05000,2.333
28,mlp.gate_proj,0.0001679264,0.05000,2.360
28,mlp.down_proj,0.0001937283,0.05000,2.891
29,self_attn.qkv_proj,0.0002001585,0.05000,1.004
29,self_attn.o_proj,0.0000316542,0.05000,0.983
29,mlp.up_proj,0.0001901275,0.05000,2.192
29,mlp.gate_proj,0.0001900747,0.05000,2.212
29,mlp.down_proj,0.0003015468,0.05000,2.929
30,self_attn.qkv_proj,0.0001382461,0.05000,1.006
30,self_attn.o_proj,0.0000314998,0.05000,1.010
30,mlp.gate_proj,0.0002096816,0.05000,2.345
30,mlp.up_proj,0.0001791842,0.05000,2.372
30,mlp.down_proj,0.0031376772,0.05000,2.968
31,self_attn.qkv_proj,0.0000025484,0.05000,1.065
31,self_attn.o_proj,0.0000041746,0.05000,1.192
31,mlp.gate_proj,0.0000727343,0.05000,2.339
31,mlp.up_proj,0.0000396456,0.05000,2.364
31,mlp.down_proj,0.0004075643,0.05000,2.979