Add files using upload-large-folder tool
Browse files- quant_log.csv +191 -191
quant_log.csv
CHANGED
@@ -1,193 +1,193 @@
|
|
1 |
layer,module,loss,samples,damp,time
|
2 |
-
0,self_attn.k_proj,0.00415716,0.01000,2.
|
3 |
-
0,self_attn.v_proj,0.00174885,0.01000,
|
4 |
-
0,self_attn.q_proj,0.00472623,0.01000,1.
|
5 |
-
0,self_attn.out_proj,0.00001367,0.01000,
|
6 |
-
0,fc1,0.00910419,0.01000,
|
7 |
-
0,fc2,0.00010223,0.01000,
|
8 |
-
1,self_attn.k_proj,0.00200916,0.01000,
|
9 |
-
1,self_attn.v_proj,0.00045392,0.01000,
|
10 |
-
1,self_attn.q_proj,0.00168526,0.01000,
|
11 |
-
1,self_attn.out_proj,0.00000113,0.01000,1.
|
12 |
-
1,fc1,0.00852806,0.01000,2.
|
13 |
-
1,fc2,0.00014238,0.01000,9.
|
14 |
-
2,self_attn.k_proj,0.00224172,0.01000,1.
|
15 |
-
2,self_attn.v_proj,0.00065303,0.01000,
|
16 |
-
2,self_attn.q_proj,0.00196235,0.01000,1.
|
17 |
-
2,self_attn.out_proj,0.00000128,0.01000,
|
18 |
-
2,fc1,0.00893701,0.01000,2.
|
19 |
-
2,fc2,0.00124839,0.01000,
|
20 |
-
3,self_attn.k_proj,0.00230825,0.01000,1.
|
21 |
-
3,self_attn.v_proj,0.00080935,0.01000,2.
|
22 |
-
3,self_attn.q_proj,0.00235409,0.01000,
|
23 |
-
3,self_attn.out_proj,0.00000197,0.01000,
|
24 |
-
3,fc1,0.00869071,0.01000,2.
|
25 |
-
3,fc2,0.00007992,0.01000,9.
|
26 |
-
4,self_attn.k_proj,0.00175245,0.01000,
|
27 |
-
4,self_attn.v_proj,0.00085283,0.01000,
|
28 |
-
4,self_attn.q_proj,0.00182772,0.01000,1.
|
29 |
-
4,self_attn.out_proj,0.00000188,0.01000,1.
|
30 |
-
4,fc1,0.00802195,0.01000,2.
|
31 |
-
4,fc2,0.00007239,0.01000,9.
|
32 |
-
5,self_attn.k_proj,0.00248873,0.01000,
|
33 |
-
5,self_attn.v_proj,0.00095311,0.01000,1.
|
34 |
-
5,self_attn.q_proj,0.00235212,0.01000,
|
35 |
-
5,self_attn.out_proj,0.00000236,0.01000,1.
|
36 |
-
5,fc1,0.00801867,0.01000,2.
|
37 |
-
5,fc2,0.00007425,0.01000,9.
|
38 |
-
6,self_attn.k_proj,0.00274498,0.01000,
|
39 |
-
6,self_attn.v_proj,0.00101407,0.01000,1.
|
40 |
6,self_attn.q_proj,0.00244245,0.01000,1.986
|
41 |
-
6,self_attn.out_proj,0.00000272,0.01000,1.
|
42 |
-
6,fc1,0.00838979,0.01000,2.
|
43 |
-
6,fc2,0.00007380,0.01000,9.
|
44 |
-
7,self_attn.k_proj,0.00328494,0.01000,
|
45 |
-
7,self_attn.v_proj,0.00108867,0.01000,
|
46 |
-
7,self_attn.q_proj,0.00277778,0.01000,
|
47 |
-
7,self_attn.out_proj,0.00000308,0.01000,
|
48 |
-
7,fc1,0.00875679,0.01000,2.
|
49 |
-
7,fc2,0.00007781,0.01000,9.
|
50 |
-
8,self_attn.k_proj,0.00421578,0.01000,1.
|
51 |
-
8,self_attn.v_proj,0.00111133,0.01000,1.
|
52 |
-
8,self_attn.q_proj,0.00319556,0.01000,1.
|
53 |
-
8,self_attn.out_proj,0.00000307,0.01000,
|
54 |
-
8,fc1,0.00934459,0.01000,2.
|
55 |
-
8,fc2,0.00007102,0.01000,9.
|
56 |
-
9,self_attn.k_proj,0.00447789,0.01000,1.
|
57 |
-
9,self_attn.v_proj,0.00117593,0.01000,
|
58 |
-
9,self_attn.q_proj,0.00340797,0.01000,2.
|
59 |
-
9,self_attn.out_proj,0.00000339,0.01000,1.
|
60 |
-
9,fc1,0.00976188,0.01000,2.
|
61 |
-
9,fc2,0.00007284,0.01000,9.
|
62 |
-
10,self_attn.k_proj,0.00497869,0.01000,1.
|
63 |
-
10,self_attn.v_proj,0.00122442,0.01000,
|
64 |
-
10,self_attn.q_proj,0.00358134,0.01000,1.
|
65 |
-
10,self_attn.out_proj,0.00000352,0.01000,
|
66 |
-
10,fc1,0.00983512,0.01000,2.
|
67 |
-
10,fc2,0.00006175,0.01000,9.
|
68 |
-
11,self_attn.k_proj,0.00482600,0.01000,1.
|
69 |
-
11,self_attn.v_proj,0.00128705,0.01000,
|
70 |
-
11,self_attn.q_proj,0.00360832,0.01000,1.
|
71 |
-
11,self_attn.out_proj,0.00000394,0.01000,1.
|
72 |
-
11,fc1,0.01003529,0.01000,2.
|
73 |
-
11,fc2,0.00005431,0.01000,9.
|
74 |
-
12,self_attn.k_proj,0.00465323,0.01000,1.
|
75 |
-
12,self_attn.v_proj,0.00143449,0.01000,
|
76 |
-
12,self_attn.q_proj,0.00370774,0.01000,1.
|
77 |
-
12,self_attn.out_proj,0.00000529,0.01000,1.
|
78 |
-
12,fc1,0.01025199,0.01000,2.
|
79 |
-
12,fc2,0.00006097,0.01000,9.
|
80 |
-
13,self_attn.k_proj,0.00485349,0.01000,1.
|
81 |
-
13,self_attn.v_proj,0.00146670,0.01000,
|
82 |
-
13,self_attn.q_proj,0.00373054,0.01000,
|
83 |
-
13,self_attn.out_proj,0.00000507,0.01000,1.
|
84 |
-
13,fc1,0.01046675,0.01000,2.
|
85 |
-
13,fc2,0.00006243,0.01000,9.
|
86 |
-
14,self_attn.k_proj,0.00459412,0.01000,1.
|
87 |
-
14,self_attn.v_proj,0.00159913,0.01000,1.
|
88 |
-
14,self_attn.q_proj,0.00372912,0.01000,
|
89 |
-
14,self_attn.out_proj,0.00000730,0.01000,1.
|
90 |
-
14,fc1,0.01084630,0.01000,2.
|
91 |
-
14,fc2,0.00011387,0.01000,
|
92 |
-
15,self_attn.k_proj,0.00486150,0.01000,1.
|
93 |
-
15,self_attn.v_proj,0.00176849,0.01000,2.
|
94 |
-
15,self_attn.q_proj,0.00400157,0.01000,1.
|
95 |
-
15,self_attn.out_proj,0.00001088,0.01000,
|
96 |
-
15,fc1,0.01212753,0.01000,2.
|
97 |
-
15,fc2,0.00018270,0.01000,9.
|
98 |
-
16,self_attn.k_proj,0.00508671,0.01000,1.
|
99 |
-
16,self_attn.v_proj,0.00204929,0.01000,1.
|
100 |
-
16,self_attn.q_proj,0.00414801,0.01000,1.
|
101 |
-
16,self_attn.out_proj,0.00001588,0.01000,1.
|
102 |
-
16,fc1,0.01262292,0.01000,2.
|
103 |
-
16,fc2,0.00014550,0.01000,
|
104 |
-
17,self_attn.k_proj,0.00514383,0.01000,1.
|
105 |
-
17,self_attn.v_proj,0.00225346,0.01000,1.
|
106 |
-
17,self_attn.q_proj,0.00418070,0.01000,1.
|
107 |
-
17,self_attn.out_proj,0.00002024,0.01000,1.
|
108 |
-
17,fc1,0.01363405,0.01000,2.
|
109 |
-
17,fc2,0.00020455,0.01000,
|
110 |
-
18,self_attn.k_proj,0.00473771,0.01000,2.
|
111 |
-
18,self_attn.v_proj,0.00262717,0.01000,1.
|
112 |
-
18,self_attn.q_proj,0.00398133,0.01000,
|
113 |
-
18,self_attn.out_proj,0.00002599,0.01000,
|
114 |
-
18,fc1,0.01541762,0.01000,2.
|
115 |
-
18,fc2,0.00026202,0.01000,9.
|
116 |
-
19,self_attn.k_proj,0.00552300,0.01000,1.
|
117 |
-
19,self_attn.v_proj,0.00305858,0.01000,1.
|
118 |
-
19,self_attn.q_proj,0.00456941,0.01000,
|
119 |
-
19,self_attn.out_proj,0.00003694,0.01000,1.
|
120 |
-
19,fc1,0.01743086,0.01000,2.
|
121 |
-
19,fc2,0.00048653,0.01000,
|
122 |
-
20,self_attn.k_proj,0.00588040,0.01000,
|
123 |
-
20,self_attn.v_proj,0.00411438,0.01000,1.
|
124 |
-
20,self_attn.q_proj,0.00500509,0.01000,1.
|
125 |
-
20,self_attn.out_proj,0.00007221,0.01000,
|
126 |
-
20,fc1,0.02207249,0.01000,2.
|
127 |
-
20,fc2,0.00057887,0.01000,9.
|
128 |
-
21,self_attn.k_proj,0.00649713,0.01000,1.
|
129 |
-
21,self_attn.v_proj,0.00491954,0.01000,1.
|
130 |
-
21,self_attn.q_proj,0.00552504,0.01000,
|
131 |
-
21,self_attn.out_proj,0.00010384,0.01000,
|
132 |
-
21,fc1,0.02501320,0.01000,2.
|
133 |
-
21,fc2,0.00093315,0.01000,9.
|
134 |
-
22,self_attn.k_proj,0.00647827,0.01000,1.
|
135 |
-
22,self_attn.v_proj,0.00537022,0.01000,2.
|
136 |
-
22,self_attn.q_proj,0.00560591,0.01000,1.
|
137 |
-
22,self_attn.out_proj,0.00015826,0.01000,1.
|
138 |
-
22,fc1,0.02930729,0.01000,2.
|
139 |
-
22,fc2,0.00103986,0.01000,9.
|
140 |
-
23,self_attn.k_proj,0.00666023,0.01000,1.
|
141 |
-
23,self_attn.v_proj,0.00584571,0.01000,1.
|
142 |
-
23,self_attn.q_proj,0.00566681,0.01000,
|
143 |
-
23,self_attn.out_proj,0.00020504,0.01000,1.
|
144 |
-
23,fc1,0.03223705,0.01000,2.
|
145 |
-
23,fc2,0.00119460,0.01000,9.
|
146 |
-
24,self_attn.k_proj,0.00652704,0.01000,
|
147 |
-
24,self_attn.v_proj,0.00684727,0.01000,
|
148 |
-
24,self_attn.q_proj,0.00548726,0.01000,
|
149 |
-
24,self_attn.out_proj,0.00028135,0.01000,1.
|
150 |
-
24,fc1,0.03299303,0.01000,2.
|
151 |
-
24,fc2,0.00129423,0.01000,9.
|
152 |
-
25,self_attn.k_proj,0.00584573,0.01000,1.
|
153 |
-
25,self_attn.v_proj,0.00735175,0.01000,1.
|
154 |
-
25,self_attn.q_proj,0.00532963,0.01000,2.
|
155 |
-
25,self_attn.out_proj,0.00040339,0.01000,1.
|
156 |
-
25,fc1,0.03120277,0.01000,2.
|
157 |
-
25,fc2,0.00092968,0.01000,9.
|
158 |
-
26,self_attn.k_proj,0.00514829,0.01000,1.
|
159 |
-
26,self_attn.v_proj,0.00683620,0.01000,1.
|
160 |
-
26,self_attn.q_proj,0.00465109,0.01000,1.
|
161 |
-
26,self_attn.out_proj,0.00038590,0.01000,1.
|
162 |
-
26,fc1,0.02892651,0.01000,2.
|
163 |
-
26,fc2,0.00102450,0.01000,
|
164 |
-
27,self_attn.k_proj,0.00492773,0.01000,1.
|
165 |
-
27,self_attn.v_proj,0.00641426,0.01000,1.
|
166 |
-
27,self_attn.q_proj,0.00446295,0.01000,1.
|
167 |
-
27,self_attn.out_proj,0.00038256,0.01000,1.
|
168 |
-
27,fc1,0.02730835,0.01000,2.
|
169 |
-
27,fc2,0.00097524,0.01000,9.
|
170 |
-
28,self_attn.k_proj,0.00465383,0.01000,1.
|
171 |
-
28,self_attn.v_proj,0.00659960,0.01000,
|
172 |
-
28,self_attn.q_proj,0.00452543,0.01000,1.
|
173 |
-
28,self_attn.out_proj,0.00041981,0.01000,1.
|
174 |
-
28,fc1,0.02555545,0.01000,2.
|
175 |
-
28,fc2,0.00089436,0.01000,8.
|
176 |
-
29,self_attn.k_proj,0.00454777,0.01000,1.
|
177 |
-
29,self_attn.v_proj,0.00631712,0.01000,
|
178 |
-
29,self_attn.q_proj,0.00478012,0.01000,1.
|
179 |
-
29,self_attn.out_proj,0.00042818,0.01000,1.
|
180 |
-
29,fc1,0.02416112,0.01000,2.
|
181 |
-
29,fc2,0.00095927,0.01000,9.
|
182 |
-
30,self_attn.k_proj,0.00441271,0.01000,
|
183 |
-
30,self_attn.v_proj,0.00628183,0.01000,1.
|
184 |
-
30,self_attn.q_proj,0.00481614,0.01000,1.
|
185 |
-
30,self_attn.out_proj,0.00034180,0.01000,1.
|
186 |
-
30,fc1,0.02313118,0.01000,2.
|
187 |
-
30,fc2,0.00110957,0.01000,
|
188 |
-
31,self_attn.k_proj,0.00427999,0.01000,
|
189 |
-
31,self_attn.v_proj,0.00589702,0.01000,1.
|
190 |
-
31,self_attn.q_proj,0.00494978,0.01000,1.
|
191 |
-
31,self_attn.out_proj,0.00033863,0.01000,1.
|
192 |
-
31,fc1,0.02249749,0.01000,2.
|
193 |
-
31,fc2,3.23435900,0.01250,10.
|
|
|
1 |
layer,module,loss,samples,damp,time
|
2 |
+
0,self_attn.k_proj,0.00415716,0.01000,2.281
|
3 |
+
0,self_attn.v_proj,0.00174885,0.01000,2.021
|
4 |
+
0,self_attn.q_proj,0.00472623,0.01000,1.992
|
5 |
+
0,self_attn.out_proj,0.00001367,0.01000,2.018
|
6 |
+
0,fc1,0.00910419,0.01000,2.042
|
7 |
+
0,fc2,0.00010223,0.01000,9.085
|
8 |
+
1,self_attn.k_proj,0.00200916,0.01000,2.003
|
9 |
+
1,self_attn.v_proj,0.00045392,0.01000,2.005
|
10 |
+
1,self_attn.q_proj,0.00168526,0.01000,2.029
|
11 |
+
1,self_attn.out_proj,0.00000113,0.01000,1.967
|
12 |
+
1,fc1,0.00852806,0.01000,2.092
|
13 |
+
1,fc2,0.00014238,0.01000,9.141
|
14 |
+
2,self_attn.k_proj,0.00224172,0.01000,1.977
|
15 |
+
2,self_attn.v_proj,0.00065303,0.01000,2.026
|
16 |
+
2,self_attn.q_proj,0.00196235,0.01000,1.987
|
17 |
+
2,self_attn.out_proj,0.00000128,0.01000,2.002
|
18 |
+
2,fc1,0.00893701,0.01000,2.169
|
19 |
+
2,fc2,0.00124839,0.01000,9.627
|
20 |
+
3,self_attn.k_proj,0.00230825,0.01000,1.972
|
21 |
+
3,self_attn.v_proj,0.00080935,0.01000,2.132
|
22 |
+
3,self_attn.q_proj,0.00235409,0.01000,2.142
|
23 |
+
3,self_attn.out_proj,0.00000197,0.01000,2.281
|
24 |
+
3,fc1,0.00869071,0.01000,2.046
|
25 |
+
3,fc2,0.00007992,0.01000,9.158
|
26 |
+
4,self_attn.k_proj,0.00175245,0.01000,2.013
|
27 |
+
4,self_attn.v_proj,0.00085283,0.01000,2.001
|
28 |
+
4,self_attn.q_proj,0.00182772,0.01000,1.979
|
29 |
+
4,self_attn.out_proj,0.00000188,0.01000,1.978
|
30 |
+
4,fc1,0.00802195,0.01000,2.065
|
31 |
+
4,fc2,0.00007239,0.01000,9.098
|
32 |
+
5,self_attn.k_proj,0.00248873,0.01000,2.040
|
33 |
+
5,self_attn.v_proj,0.00095311,0.01000,1.995
|
34 |
+
5,self_attn.q_proj,0.00235212,0.01000,2.021
|
35 |
+
5,self_attn.out_proj,0.00000236,0.01000,1.963
|
36 |
+
5,fc1,0.00801867,0.01000,2.090
|
37 |
+
5,fc2,0.00007425,0.01000,9.223
|
38 |
+
6,self_attn.k_proj,0.00274498,0.01000,2.011
|
39 |
+
6,self_attn.v_proj,0.00101407,0.01000,1.997
|
40 |
6,self_attn.q_proj,0.00244245,0.01000,1.986
|
41 |
+
6,self_attn.out_proj,0.00000272,0.01000,1.977
|
42 |
+
6,fc1,0.00838979,0.01000,2.041
|
43 |
+
6,fc2,0.00007380,0.01000,9.011
|
44 |
+
7,self_attn.k_proj,0.00328494,0.01000,1.965
|
45 |
+
7,self_attn.v_proj,0.00108867,0.01000,2.001
|
46 |
+
7,self_attn.q_proj,0.00277778,0.01000,1.991
|
47 |
+
7,self_attn.out_proj,0.00000308,0.01000,2.023
|
48 |
+
7,fc1,0.00875679,0.01000,2.076
|
49 |
+
7,fc2,0.00007781,0.01000,9.106
|
50 |
+
8,self_attn.k_proj,0.00421578,0.01000,1.960
|
51 |
+
8,self_attn.v_proj,0.00111133,0.01000,1.983
|
52 |
+
8,self_attn.q_proj,0.00319556,0.01000,1.963
|
53 |
+
8,self_attn.out_proj,0.00000307,0.01000,1.988
|
54 |
+
8,fc1,0.00934459,0.01000,2.063
|
55 |
+
8,fc2,0.00007102,0.01000,9.116
|
56 |
+
9,self_attn.k_proj,0.00447789,0.01000,1.991
|
57 |
+
9,self_attn.v_proj,0.00117593,0.01000,2.002
|
58 |
+
9,self_attn.q_proj,0.00340797,0.01000,2.044
|
59 |
+
9,self_attn.out_proj,0.00000339,0.01000,1.999
|
60 |
+
9,fc1,0.00976188,0.01000,2.074
|
61 |
+
9,fc2,0.00007284,0.01000,9.097
|
62 |
+
10,self_attn.k_proj,0.00497869,0.01000,1.979
|
63 |
+
10,self_attn.v_proj,0.00122442,0.01000,2.010
|
64 |
+
10,self_attn.q_proj,0.00358134,0.01000,1.973
|
65 |
+
10,self_attn.out_proj,0.00000352,0.01000,2.006
|
66 |
+
10,fc1,0.00983512,0.01000,2.066
|
67 |
+
10,fc2,0.00006175,0.01000,9.069
|
68 |
+
11,self_attn.k_proj,0.00482600,0.01000,1.974
|
69 |
+
11,self_attn.v_proj,0.00128705,0.01000,1.997
|
70 |
+
11,self_attn.q_proj,0.00360832,0.01000,1.978
|
71 |
+
11,self_attn.out_proj,0.00000394,0.01000,1.985
|
72 |
+
11,fc1,0.01003529,0.01000,2.062
|
73 |
+
11,fc2,0.00005431,0.01000,9.185
|
74 |
+
12,self_attn.k_proj,0.00465323,0.01000,1.978
|
75 |
+
12,self_attn.v_proj,0.00143449,0.01000,1.989
|
76 |
+
12,self_attn.q_proj,0.00370774,0.01000,1.992
|
77 |
+
12,self_attn.out_proj,0.00000529,0.01000,1.965
|
78 |
+
12,fc1,0.01025199,0.01000,2.059
|
79 |
+
12,fc2,0.00006097,0.01000,9.127
|
80 |
+
13,self_attn.k_proj,0.00485349,0.01000,1.967
|
81 |
+
13,self_attn.v_proj,0.00146670,0.01000,2.011
|
82 |
+
13,self_attn.q_proj,0.00373054,0.01000,2.071
|
83 |
+
13,self_attn.out_proj,0.00000507,0.01000,1.949
|
84 |
+
13,fc1,0.01046675,0.01000,2.054
|
85 |
+
13,fc2,0.00006243,0.01000,9.087
|
86 |
+
14,self_attn.k_proj,0.00459412,0.01000,1.948
|
87 |
+
14,self_attn.v_proj,0.00159913,0.01000,1.970
|
88 |
+
14,self_attn.q_proj,0.00372912,0.01000,2.049
|
89 |
+
14,self_attn.out_proj,0.00000730,0.01000,1.978
|
90 |
+
14,fc1,0.01084630,0.01000,2.063
|
91 |
+
14,fc2,0.00011387,0.01000,9.671
|
92 |
+
15,self_attn.k_proj,0.00486150,0.01000,1.964
|
93 |
+
15,self_attn.v_proj,0.00176849,0.01000,2.215
|
94 |
+
15,self_attn.q_proj,0.00400157,0.01000,1.992
|
95 |
+
15,self_attn.out_proj,0.00001088,0.01000,2.161
|
96 |
+
15,fc1,0.01212753,0.01000,2.063
|
97 |
+
15,fc2,0.00018270,0.01000,9.106
|
98 |
+
16,self_attn.k_proj,0.00508671,0.01000,1.960
|
99 |
+
16,self_attn.v_proj,0.00204929,0.01000,1.995
|
100 |
+
16,self_attn.q_proj,0.00414801,0.01000,1.961
|
101 |
+
16,self_attn.out_proj,0.00001588,0.01000,1.985
|
102 |
+
16,fc1,0.01262292,0.01000,2.057
|
103 |
+
16,fc2,0.00014550,0.01000,9.127
|
104 |
+
17,self_attn.k_proj,0.00514383,0.01000,1.962
|
105 |
+
17,self_attn.v_proj,0.00225346,0.01000,1.984
|
106 |
+
17,self_attn.q_proj,0.00418070,0.01000,1.966
|
107 |
+
17,self_attn.out_proj,0.00002024,0.01000,1.991
|
108 |
+
17,fc1,0.01363405,0.01000,2.164
|
109 |
+
17,fc2,0.00020455,0.01000,9.124
|
110 |
+
18,self_attn.k_proj,0.00473771,0.01000,2.006
|
111 |
+
18,self_attn.v_proj,0.00262717,0.01000,1.966
|
112 |
+
18,self_attn.q_proj,0.00398133,0.01000,1.987
|
113 |
+
18,self_attn.out_proj,0.00002599,0.01000,1.956
|
114 |
+
18,fc1,0.01541762,0.01000,2.037
|
115 |
+
18,fc2,0.00026202,0.01000,9.153
|
116 |
+
19,self_attn.k_proj,0.00552300,0.01000,1.995
|
117 |
+
19,self_attn.v_proj,0.00305858,0.01000,1.979
|
118 |
+
19,self_attn.q_proj,0.00456941,0.01000,2.011
|
119 |
+
19,self_attn.out_proj,0.00003694,0.01000,1.950
|
120 |
+
19,fc1,0.01743086,0.01000,2.064
|
121 |
+
19,fc2,0.00048653,0.01000,9.156
|
122 |
+
20,self_attn.k_proj,0.00588040,0.01000,2.004
|
123 |
+
20,self_attn.v_proj,0.00411438,0.01000,1.997
|
124 |
+
20,self_attn.q_proj,0.00500509,0.01000,1.960
|
125 |
+
20,self_attn.out_proj,0.00007221,0.01000,1.971
|
126 |
+
20,fc1,0.02207249,0.01000,2.068
|
127 |
+
20,fc2,0.00057887,0.01000,9.064
|
128 |
+
21,self_attn.k_proj,0.00649713,0.01000,1.963
|
129 |
+
21,self_attn.v_proj,0.00491954,0.01000,1.992
|
130 |
+
21,self_attn.q_proj,0.00552504,0.01000,1.970
|
131 |
+
21,self_attn.out_proj,0.00010384,0.01000,2.003
|
132 |
+
21,fc1,0.02501320,0.01000,2.053
|
133 |
+
21,fc2,0.00093315,0.01000,9.096
|
134 |
+
22,self_attn.k_proj,0.00647827,0.01000,1.960
|
135 |
+
22,self_attn.v_proj,0.00537022,0.01000,2.011
|
136 |
+
22,self_attn.q_proj,0.00560591,0.01000,1.990
|
137 |
+
22,self_attn.out_proj,0.00015826,0.01000,1.979
|
138 |
+
22,fc1,0.02930729,0.01000,2.096
|
139 |
+
22,fc2,0.00103986,0.01000,9.092
|
140 |
+
23,self_attn.k_proj,0.00666023,0.01000,1.972
|
141 |
+
23,self_attn.v_proj,0.00584571,0.01000,1.964
|
142 |
+
23,self_attn.q_proj,0.00566681,0.01000,1.991
|
143 |
+
23,self_attn.out_proj,0.00020504,0.01000,1.972
|
144 |
+
23,fc1,0.03223705,0.01000,2.057
|
145 |
+
23,fc2,0.00119460,0.01000,9.091
|
146 |
+
24,self_attn.k_proj,0.00652704,0.01000,2.000
|
147 |
+
24,self_attn.v_proj,0.00684727,0.01000,1.978
|
148 |
+
24,self_attn.q_proj,0.00548726,0.01000,1.996
|
149 |
+
24,self_attn.out_proj,0.00028135,0.01000,1.975
|
150 |
+
24,fc1,0.03299303,0.01000,2.082
|
151 |
+
24,fc2,0.00129423,0.01000,9.043
|
152 |
+
25,self_attn.k_proj,0.00584573,0.01000,1.965
|
153 |
+
25,self_attn.v_proj,0.00735175,0.01000,1.994
|
154 |
+
25,self_attn.q_proj,0.00532963,0.01000,2.012
|
155 |
+
25,self_attn.out_proj,0.00040339,0.01000,1.974
|
156 |
+
25,fc1,0.03120277,0.01000,2.076
|
157 |
+
25,fc2,0.00092968,0.01000,9.085
|
158 |
+
26,self_attn.k_proj,0.00514829,0.01000,1.958
|
159 |
+
26,self_attn.v_proj,0.00683620,0.01000,1.957
|
160 |
+
26,self_attn.q_proj,0.00465109,0.01000,1.981
|
161 |
+
26,self_attn.out_proj,0.00038590,0.01000,1.946
|
162 |
+
26,fc1,0.02892651,0.01000,2.058
|
163 |
+
26,fc2,0.00102450,0.01000,9.106
|
164 |
+
27,self_attn.k_proj,0.00492773,0.01000,1.980
|
165 |
+
27,self_attn.v_proj,0.00641426,0.01000,1.986
|
166 |
+
27,self_attn.q_proj,0.00446295,0.01000,1.978
|
167 |
+
27,self_attn.out_proj,0.00038256,0.01000,1.982
|
168 |
+
27,fc1,0.02730835,0.01000,2.058
|
169 |
+
27,fc2,0.00097524,0.01000,9.100
|
170 |
+
28,self_attn.k_proj,0.00465383,0.01000,1.982
|
171 |
+
28,self_attn.v_proj,0.00659960,0.01000,1.988
|
172 |
+
28,self_attn.q_proj,0.00452543,0.01000,1.967
|
173 |
+
28,self_attn.out_proj,0.00041981,0.01000,1.944
|
174 |
+
28,fc1,0.02555545,0.01000,2.051
|
175 |
+
28,fc2,0.00089436,0.01000,8.963
|
176 |
+
29,self_attn.k_proj,0.00454777,0.01000,1.961
|
177 |
+
29,self_attn.v_proj,0.00631712,0.01000,1.983
|
178 |
+
29,self_attn.q_proj,0.00478012,0.01000,1.970
|
179 |
+
29,self_attn.out_proj,0.00042818,0.01000,1.946
|
180 |
+
29,fc1,0.02416112,0.01000,2.075
|
181 |
+
29,fc2,0.00095927,0.01000,9.127
|
182 |
+
30,self_attn.k_proj,0.00441271,0.01000,2.008
|
183 |
+
30,self_attn.v_proj,0.00628183,0.01000,1.987
|
184 |
+
30,self_attn.q_proj,0.00481614,0.01000,1.971
|
185 |
+
30,self_attn.out_proj,0.00034180,0.01000,1.977
|
186 |
+
30,fc1,0.02313118,0.01000,2.037
|
187 |
+
30,fc2,0.00110957,0.01000,9.063
|
188 |
+
31,self_attn.k_proj,0.00427999,0.01000,2.017
|
189 |
+
31,self_attn.v_proj,0.00589702,0.01000,1.971
|
190 |
+
31,self_attn.q_proj,0.00494978,0.01000,1.962
|
191 |
+
31,self_attn.out_proj,0.00033863,0.01000,1.956
|
192 |
+
31,fc1,0.02249749,0.01000,2.093
|
193 |
+
31,fc2,3.23435900,0.01250,10.112
|