dmeck's picture
Upload folder using huggingface_hub
24ef7dc verified
layer,module,loss,samples,damp,time
0,self_attn.k_proj,0.0000001474,0.01250,2.927
0,self_attn.v_proj,0.0000000727,0.01250,1.000
0,self_attn.q_proj,0.0000012972,0.01250,1.009
0,self_attn.o_proj,0.0000006876,0.01000,0.981
0,mlp.gate_up_proj,0.0000313005,0.01000,1.050
0,mlp.down_proj,0.0000222758,0.01000,3.867
1,self_attn.k_proj,0.0000001293,0.01000,0.959
1,self_attn.v_proj,0.0000000880,0.01000,0.953
1,self_attn.q_proj,0.0000014647,0.01000,0.971
1,self_attn.o_proj,0.0000010034,0.01000,0.963
1,mlp.gate_up_proj,0.0000223667,0.01000,1.019
1,mlp.down_proj,0.0000057445,0.01000,3.856
2,self_attn.k_proj,0.0000001341,0.01000,0.973
2,self_attn.v_proj,0.0000000936,0.01000,0.990
2,self_attn.q_proj,0.0000014909,0.01000,0.997
2,self_attn.o_proj,0.0000007911,0.01000,0.982
2,mlp.gate_up_proj,0.0000266103,0.01000,1.044
2,mlp.down_proj,0.0000048187,0.01000,3.888
3,self_attn.k_proj,0.0000001288,0.01000,0.956
3,self_attn.v_proj,0.0000001014,0.01000,0.953
3,self_attn.q_proj,0.0000015294,0.01000,0.976
3,self_attn.o_proj,0.0000003380,0.01000,0.966
3,mlp.gate_up_proj,0.0000243513,0.01000,1.009
3,mlp.down_proj,0.0000036748,0.01000,3.841
4,self_attn.k_proj,0.0000001223,0.01000,0.975
4,self_attn.v_proj,0.0000000823,0.01000,0.961
4,self_attn.q_proj,0.0000013132,0.01000,0.989
4,self_attn.o_proj,0.0000002347,0.01000,0.965
4,mlp.gate_up_proj,0.0000171597,0.01000,1.024
4,mlp.down_proj,0.0000020520,0.01000,3.939
5,self_attn.k_proj,0.0000002609,0.01000,0.952
5,self_attn.v_proj,0.0000002149,0.01000,0.947
5,self_attn.q_proj,0.0000031802,0.01000,0.967
5,self_attn.o_proj,0.0000003969,0.01000,0.959
5,mlp.gate_up_proj,0.0000133494,0.01000,1.018
5,mlp.down_proj,0.0000010682,0.01000,3.864
6,self_attn.k_proj,0.0000001916,0.01000,0.952
6,self_attn.v_proj,0.0000001461,0.01000,0.947
6,self_attn.q_proj,0.0000022685,0.01000,0.966
6,self_attn.o_proj,0.0000002764,0.01000,0.967
6,mlp.gate_up_proj,0.0000116535,0.01000,1.017
6,mlp.down_proj,0.0000007555,0.01000,3.934
7,self_attn.k_proj,0.0000001383,0.01000,0.956
7,self_attn.v_proj,0.0000001067,0.01000,0.950
7,self_attn.q_proj,0.0000017731,0.01000,0.967
7,self_attn.o_proj,0.0000001694,0.01000,0.976
7,mlp.gate_up_proj,0.0000112911,0.01000,1.022
7,mlp.down_proj,0.0000007493,0.01000,3.865
8,self_attn.k_proj,0.0000002279,0.01000,0.957
8,self_attn.v_proj,0.0000001769,0.01000,0.949
8,self_attn.q_proj,0.0000028795,0.01000,0.963
8,self_attn.o_proj,0.0000003606,0.01000,0.957
8,mlp.gate_up_proj,0.0000120715,0.01000,1.019
8,mlp.down_proj,0.0000007933,0.01000,3.906
9,self_attn.k_proj,0.0000002795,0.01000,0.959
9,self_attn.v_proj,0.0000001922,0.01000,0.952
9,self_attn.q_proj,0.0000029595,0.01000,0.973
9,self_attn.o_proj,0.0000003512,0.01000,0.963
9,mlp.gate_up_proj,0.0000115193,0.01000,1.010
9,mlp.down_proj,0.0000006440,0.01000,3.884
10,self_attn.k_proj,0.0000002780,0.01000,0.954
10,self_attn.v_proj,0.0000002080,0.01000,0.951
10,self_attn.q_proj,0.0000031246,0.01000,0.956
10,self_attn.o_proj,0.0000001547,0.01000,0.957
10,mlp.gate_up_proj,0.0000119229,0.01000,1.022
10,mlp.down_proj,0.0000005447,0.01000,3.881
11,self_attn.k_proj,0.0000003099,0.01000,0.954
11,self_attn.v_proj,0.0000002167,0.01000,0.953
11,self_attn.q_proj,0.0000034380,0.01000,0.971
11,self_attn.o_proj,0.0000003469,0.01000,0.966
11,mlp.gate_up_proj,0.0000120436,0.01000,1.022
11,mlp.down_proj,0.0000005383,0.01000,3.869
12,self_attn.k_proj,0.0000003269,0.01000,0.954
12,self_attn.v_proj,0.0000002273,0.01000,0.949
12,self_attn.q_proj,0.0000037323,0.01000,0.970
12,self_attn.o_proj,0.0000002915,0.01000,0.969
12,mlp.gate_up_proj,0.0000124333,0.01000,1.028
12,mlp.down_proj,0.0000005790,0.01000,3.892
13,self_attn.k_proj,0.0000002875,0.01000,0.961
13,self_attn.v_proj,0.0000002054,0.01000,0.957
13,self_attn.q_proj,0.0000032311,0.01000,0.974
13,self_attn.o_proj,0.0000004909,0.01000,0.963
13,mlp.gate_up_proj,0.0000109160,0.01000,1.009
13,mlp.down_proj,0.0000004969,0.01000,3.892
14,self_attn.k_proj,0.0000002173,0.01000,0.947
14,self_attn.v_proj,0.0000001522,0.01000,0.946
14,self_attn.q_proj,0.0000025553,0.01000,0.963
14,self_attn.o_proj,0.0000003246,0.01000,0.971
14,mlp.gate_up_proj,0.0000097154,0.01000,1.024
14,mlp.down_proj,0.0000004670,0.01000,3.859
15,self_attn.k_proj,0.0000003722,0.01000,0.952
15,self_attn.v_proj,0.0000002349,0.01000,0.950
15,self_attn.q_proj,0.0000038035,0.01000,0.968
15,self_attn.o_proj,0.0000005418,0.01000,0.972
15,mlp.gate_up_proj,0.0000093693,0.01000,1.029
15,mlp.down_proj,0.0000004739,0.01000,3.943
16,self_attn.k_proj,0.0000002379,0.01000,0.971
16,self_attn.v_proj,0.0000001854,0.01000,0.974
16,self_attn.q_proj,0.0000028935,0.01000,0.995
16,self_attn.o_proj,0.0000003674,0.01000,0.960
16,mlp.gate_up_proj,0.0000074566,0.01000,1.021
16,mlp.down_proj,0.0000002783,0.01000,3.904
17,self_attn.k_proj,0.0000003128,0.01000,0.964
17,self_attn.v_proj,0.0000002831,0.01000,0.964
17,self_attn.q_proj,0.0000046325,0.01000,0.974
17,self_attn.o_proj,0.0000005550,0.01000,0.998
17,mlp.gate_up_proj,0.0000069269,0.01000,1.020
17,mlp.down_proj,0.0000002150,0.01000,3.906
18,self_attn.k_proj,0.0000002372,0.01000,0.952
18,self_attn.v_proj,0.0000002249,0.01000,0.955
18,self_attn.q_proj,0.0000034291,0.01000,0.971
18,self_attn.o_proj,0.0000003502,0.01000,0.950
18,mlp.gate_up_proj,0.0000065477,0.01000,1.014
18,mlp.down_proj,0.0000001969,0.01000,3.903
19,self_attn.k_proj,0.0000002907,0.01000,0.958
19,self_attn.v_proj,0.0000002680,0.01000,0.950
19,self_attn.q_proj,0.0000040165,0.01000,0.966
19,self_attn.o_proj,0.0000003714,0.01000,0.962
19,mlp.gate_up_proj,0.0000041124,0.01000,1.021
19,mlp.down_proj,0.0000000854,0.01000,3.902
20,self_attn.k_proj,0.0000003351,0.01000,0.956
20,self_attn.v_proj,0.0000003156,0.01000,0.952
20,self_attn.q_proj,0.0000045689,0.01000,0.973
20,self_attn.o_proj,0.0000004012,0.01000,0.965
20,mlp.gate_up_proj,0.0000064211,0.01000,1.015
20,mlp.down_proj,0.0000001780,0.01000,3.881
21,self_attn.k_proj,0.0000002523,0.01000,0.970
21,self_attn.v_proj,0.0000002327,0.01000,0.963
21,self_attn.q_proj,0.0000038044,0.01000,0.975
21,self_attn.o_proj,0.0000003798,0.01000,0.961
21,mlp.gate_up_proj,0.0000060094,0.01000,1.023
21,mlp.down_proj,0.0000001864,0.01000,3.879
22,self_attn.k_proj,0.0000002464,0.01000,0.956
22,self_attn.v_proj,0.0000002398,0.01000,0.990
22,self_attn.q_proj,0.0000038630,0.01000,0.983
22,self_attn.o_proj,0.0000004510,0.01000,0.969
22,mlp.gate_up_proj,0.0000064875,0.01000,1.020
22,mlp.down_proj,0.0000002224,0.01000,3.849
23,self_attn.k_proj,0.0000003366,0.01000,0.959
23,self_attn.v_proj,0.0000003180,0.01000,0.951
23,self_attn.q_proj,0.0000046087,0.01000,0.968
23,self_attn.o_proj,0.0000003156,0.01000,0.967
23,mlp.gate_up_proj,0.0000067177,0.01000,1.012
23,mlp.down_proj,0.0000002321,0.01000,3.867
24,self_attn.k_proj,0.0000003049,0.01000,0.967
24,self_attn.v_proj,0.0000002931,0.01000,0.976
24,self_attn.q_proj,0.0000041782,0.01000,0.978
24,self_attn.o_proj,0.0000003220,0.01000,0.980
24,mlp.gate_up_proj,0.0000067377,0.01000,1.040
24,mlp.down_proj,0.0000002455,0.01000,3.918
25,self_attn.k_proj,0.0000002532,0.01000,0.965
25,self_attn.v_proj,0.0000002051,0.01000,0.979
25,self_attn.q_proj,0.0000038520,0.01000,0.996
25,self_attn.o_proj,0.0000003924,0.01000,0.967
25,mlp.gate_up_proj,0.0000063683,0.01000,1.018
25,mlp.down_proj,0.0000002687,0.01000,3.937
26,self_attn.k_proj,0.0000002828,0.01000,0.965
26,self_attn.v_proj,0.0000002964,0.01000,0.963
26,self_attn.q_proj,0.0000038521,0.01000,0.980
26,self_attn.o_proj,0.0000003255,0.01000,0.971
26,mlp.gate_up_proj,0.0000065724,0.01000,1.021
26,mlp.down_proj,0.0000002814,0.01000,3.866
27,self_attn.k_proj,0.0000003111,0.01000,0.962
27,self_attn.v_proj,0.0000003142,0.01000,0.959
27,self_attn.q_proj,0.0000044170,0.01000,0.975
27,self_attn.o_proj,0.0000001249,0.01000,0.973
27,mlp.gate_up_proj,0.0000061170,0.01000,1.020
27,mlp.down_proj,0.0000002248,0.01000,3.926
28,self_attn.k_proj,0.0000002896,0.01000,0.954
28,self_attn.v_proj,0.0000002210,0.01000,0.961
28,self_attn.q_proj,0.0000033302,0.01000,0.967
28,self_attn.o_proj,0.0000001354,0.01000,0.962
28,mlp.gate_up_proj,0.0000072945,0.01000,1.025
28,mlp.down_proj,0.0000002744,0.01000,3.864
29,self_attn.k_proj,0.0000003282,0.01000,0.953
29,self_attn.v_proj,0.0000002764,0.01000,0.953
29,self_attn.q_proj,0.0000036271,0.01000,0.974
29,self_attn.o_proj,0.0000000919,0.01000,0.963
29,mlp.gate_up_proj,0.0000067340,0.01000,1.012
29,mlp.down_proj,0.0000002345,0.01000,3.893
30,self_attn.k_proj,0.0000003301,0.01000,0.964
30,self_attn.v_proj,0.0000002453,0.01000,0.953
30,self_attn.q_proj,0.0000039882,0.01000,0.969
30,self_attn.o_proj,0.0000000933,0.01000,0.964
30,mlp.gate_up_proj,0.0000056270,0.01000,1.003
30,mlp.down_proj,0.0000001861,0.01000,3.894
31,self_attn.k_proj,0.0000003153,0.01000,0.955
31,self_attn.v_proj,0.0000002457,0.01000,0.950
31,self_attn.q_proj,0.0000036933,0.01000,0.967
31,self_attn.o_proj,0.0000001759,0.01000,0.976
31,mlp.gate_up_proj,0.0000049036,0.01000,1.037
31,mlp.down_proj,0.0000001383,0.01000,3.967
32,self_attn.k_proj,0.0000003170,0.01000,0.968
32,self_attn.v_proj,0.0000002645,0.01000,0.954
32,self_attn.q_proj,0.0000033426,0.01000,0.994
32,self_attn.o_proj,0.0000000663,0.01000,0.984
32,mlp.gate_up_proj,0.0000046289,0.01000,1.045
32,mlp.down_proj,0.0000001111,0.01000,3.896
33,self_attn.k_proj,0.0000003068,0.01000,0.982
33,self_attn.v_proj,0.0000002334,0.01000,0.978
33,self_attn.q_proj,0.0000031778,0.01000,0.994
33,self_attn.o_proj,0.0000000625,0.01000,0.968
33,mlp.gate_up_proj,0.0000044431,0.01000,1.029
33,mlp.down_proj,0.0000001145,0.01000,3.872
34,self_attn.k_proj,0.0000003057,0.01000,0.963
34,self_attn.v_proj,0.0000002110,0.01000,0.957
34,self_attn.q_proj,0.0000033824,0.01000,0.986
34,self_attn.o_proj,0.0000003696,0.01000,0.962
34,mlp.gate_up_proj,0.0000046237,0.01000,1.023
34,mlp.down_proj,0.0000001118,0.01000,3.888
35,self_attn.k_proj,0.0000003121,0.01000,0.959
35,self_attn.v_proj,0.0000002413,0.01000,0.954
35,self_attn.q_proj,0.0000036854,0.01000,0.969
35,self_attn.o_proj,0.0000000996,0.01000,0.963
35,mlp.gate_up_proj,0.0000062732,0.01000,1.019
35,mlp.down_proj,0.0000002262,0.01000,3.895
36,self_attn.k_proj,0.0000002683,0.01000,0.974
36,self_attn.v_proj,0.0000001936,0.01000,0.969
36,self_attn.q_proj,0.0000029817,0.01000,0.985
36,self_attn.o_proj,0.0000001425,0.01000,0.982
36,mlp.gate_up_proj,0.0000079659,0.01000,1.013
36,mlp.down_proj,0.0000003427,0.01000,3.874
37,self_attn.k_proj,0.0000002539,0.01000,0.957
37,self_attn.v_proj,0.0000001903,0.01000,0.953
37,self_attn.q_proj,0.0000030935,0.01000,0.974
37,self_attn.o_proj,0.0000001804,0.01000,0.969
37,mlp.gate_up_proj,0.0000131213,0.01000,1.027
37,mlp.down_proj,0.0000010023,0.01000,3.854
38,self_attn.k_proj,0.0000002297,0.01000,0.961
38,self_attn.v_proj,0.0000001820,0.01000,0.957
38,self_attn.q_proj,0.0000030453,0.01000,0.974
38,self_attn.o_proj,0.0000002469,0.01000,0.968
38,mlp.gate_up_proj,0.0000168279,0.01000,1.021
38,mlp.down_proj,0.0000024667,0.01000,3.892
39,self_attn.k_proj,0.0000002316,0.01000,0.967
39,self_attn.v_proj,0.0000001737,0.01000,0.966
39,self_attn.q_proj,0.0000028477,0.01000,0.986
39,self_attn.o_proj,0.0000001884,0.01000,0.993
39,mlp.gate_up_proj,0.0000174337,0.01000,1.020
39,mlp.down_proj,0.0000020299,0.01000,3.913