Upload attention_entropies.jsonl with huggingface_hub
Browse files- attention_entropies.jsonl +144 -0
attention_entropies.jsonl
ADDED
@@ -0,0 +1,144 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{"layer":0,"head":0,"avg_attn_entropy":5.3548670359}
|
2 |
+
{"layer":0,"head":1,"avg_attn_entropy":5.4073407726}
|
3 |
+
{"layer":0,"head":2,"avg_attn_entropy":5.6437054749}
|
4 |
+
{"layer":0,"head":3,"avg_attn_entropy":5.6086924386}
|
5 |
+
{"layer":0,"head":4,"avg_attn_entropy":5.5739436259}
|
6 |
+
{"layer":0,"head":5,"avg_attn_entropy":5.8404562573}
|
7 |
+
{"layer":0,"head":6,"avg_attn_entropy":5.7293018761}
|
8 |
+
{"layer":0,"head":7,"avg_attn_entropy":5.5945004225}
|
9 |
+
{"layer":0,"head":8,"avg_attn_entropy":5.5260455632}
|
10 |
+
{"layer":0,"head":9,"avg_attn_entropy":5.0169561224}
|
11 |
+
{"layer":0,"head":10,"avg_attn_entropy":5.4916863084}
|
12 |
+
{"layer":0,"head":11,"avg_attn_entropy":4.4476501932}
|
13 |
+
{"layer":1,"head":0,"avg_attn_entropy":4.417647264}
|
14 |
+
{"layer":1,"head":1,"avg_attn_entropy":4.4385528791}
|
15 |
+
{"layer":1,"head":2,"avg_attn_entropy":5.4517773166}
|
16 |
+
{"layer":1,"head":3,"avg_attn_entropy":3.4159071567}
|
17 |
+
{"layer":1,"head":4,"avg_attn_entropy":4.9185877376}
|
18 |
+
{"layer":1,"head":5,"avg_attn_entropy":4.5399041615}
|
19 |
+
{"layer":1,"head":6,"avg_attn_entropy":2.3325746882}
|
20 |
+
{"layer":1,"head":7,"avg_attn_entropy":0.0654781002}
|
21 |
+
{"layer":1,"head":8,"avg_attn_entropy":5.6879639049}
|
22 |
+
{"layer":1,"head":9,"avg_attn_entropy":5.3854743338}
|
23 |
+
{"layer":1,"head":10,"avg_attn_entropy":4.3788061266}
|
24 |
+
{"layer":1,"head":11,"avg_attn_entropy":5.5759583545}
|
25 |
+
{"layer":2,"head":0,"avg_attn_entropy":0.0248770704}
|
26 |
+
{"layer":2,"head":1,"avg_attn_entropy":5.3834004121}
|
27 |
+
{"layer":2,"head":2,"avg_attn_entropy":4.5823903167}
|
28 |
+
{"layer":2,"head":3,"avg_attn_entropy":5.3243958998}
|
29 |
+
{"layer":2,"head":4,"avg_attn_entropy":4.281496253}
|
30 |
+
{"layer":2,"head":5,"avg_attn_entropy":5.0837480345}
|
31 |
+
{"layer":2,"head":6,"avg_attn_entropy":2.0900593653}
|
32 |
+
{"layer":2,"head":7,"avg_attn_entropy":5.3092582607}
|
33 |
+
{"layer":2,"head":8,"avg_attn_entropy":5.2302970176}
|
34 |
+
{"layer":2,"head":9,"avg_attn_entropy":4.8136256933}
|
35 |
+
{"layer":2,"head":10,"avg_attn_entropy":0.567478481}
|
36 |
+
{"layer":2,"head":11,"avg_attn_entropy":2.387462343}
|
37 |
+
{"layer":3,"head":0,"avg_attn_entropy":4.7737403023}
|
38 |
+
{"layer":3,"head":1,"avg_attn_entropy":3.3615415652}
|
39 |
+
{"layer":3,"head":2,"avg_attn_entropy":4.1151399989}
|
40 |
+
{"layer":3,"head":3,"avg_attn_entropy":4.8960805532}
|
41 |
+
{"layer":3,"head":4,"avg_attn_entropy":5.1519064598}
|
42 |
+
{"layer":3,"head":5,"avg_attn_entropy":4.2462465441}
|
43 |
+
{"layer":3,"head":6,"avg_attn_entropy":4.8886691146}
|
44 |
+
{"layer":3,"head":7,"avg_attn_entropy":5.0914810891}
|
45 |
+
{"layer":3,"head":8,"avg_attn_entropy":4.3073498609}
|
46 |
+
{"layer":3,"head":9,"avg_attn_entropy":4.4364870465}
|
47 |
+
{"layer":3,"head":10,"avg_attn_entropy":4.5778079004}
|
48 |
+
{"layer":3,"head":11,"avg_attn_entropy":4.9419590864}
|
49 |
+
{"layer":4,"head":0,"avg_attn_entropy":5.1810433729}
|
50 |
+
{"layer":4,"head":1,"avg_attn_entropy":5.219964781}
|
51 |
+
{"layer":4,"head":2,"avg_attn_entropy":5.4302420516}
|
52 |
+
{"layer":4,"head":3,"avg_attn_entropy":4.8663788891}
|
53 |
+
{"layer":4,"head":4,"avg_attn_entropy":2.903820996}
|
54 |
+
{"layer":4,"head":5,"avg_attn_entropy":4.0474161208}
|
55 |
+
{"layer":4,"head":6,"avg_attn_entropy":4.2850197306}
|
56 |
+
{"layer":4,"head":7,"avg_attn_entropy":5.2420392725}
|
57 |
+
{"layer":4,"head":8,"avg_attn_entropy":4.3524293251}
|
58 |
+
{"layer":4,"head":9,"avg_attn_entropy":4.9510341766}
|
59 |
+
{"layer":4,"head":10,"avg_attn_entropy":3.1817292917}
|
60 |
+
{"layer":4,"head":11,"avg_attn_entropy":2.4894239509}
|
61 |
+
{"layer":5,"head":0,"avg_attn_entropy":4.1062956436}
|
62 |
+
{"layer":5,"head":1,"avg_attn_entropy":4.3226871593}
|
63 |
+
{"layer":5,"head":2,"avg_attn_entropy":3.3107177069}
|
64 |
+
{"layer":5,"head":3,"avg_attn_entropy":3.9191159055}
|
65 |
+
{"layer":5,"head":4,"avg_attn_entropy":3.1964480085}
|
66 |
+
{"layer":5,"head":5,"avg_attn_entropy":0.1987101573}
|
67 |
+
{"layer":5,"head":6,"avg_attn_entropy":2.8824583101}
|
68 |
+
{"layer":5,"head":7,"avg_attn_entropy":4.146755477}
|
69 |
+
{"layer":5,"head":8,"avg_attn_entropy":2.6306450194}
|
70 |
+
{"layer":5,"head":9,"avg_attn_entropy":3.8055427697}
|
71 |
+
{"layer":5,"head":10,"avg_attn_entropy":2.8109682602}
|
72 |
+
{"layer":5,"head":11,"avg_attn_entropy":3.1027413585}
|
73 |
+
{"layer":6,"head":0,"avg_attn_entropy":3.1586957295}
|
74 |
+
{"layer":6,"head":1,"avg_attn_entropy":2.6237978191}
|
75 |
+
{"layer":6,"head":2,"avg_attn_entropy":3.687537878}
|
76 |
+
{"layer":6,"head":3,"avg_attn_entropy":3.4518884988}
|
77 |
+
{"layer":6,"head":4,"avg_attn_entropy":3.4053203999}
|
78 |
+
{"layer":6,"head":5,"avg_attn_entropy":3.7397581093}
|
79 |
+
{"layer":6,"head":6,"avg_attn_entropy":3.571864113}
|
80 |
+
{"layer":6,"head":7,"avg_attn_entropy":3.4767464311}
|
81 |
+
{"layer":6,"head":8,"avg_attn_entropy":3.7394159824}
|
82 |
+
{"layer":6,"head":9,"avg_attn_entropy":5.1220063081}
|
83 |
+
{"layer":6,"head":10,"avg_attn_entropy":3.957284929}
|
84 |
+
{"layer":6,"head":11,"avg_attn_entropy":2.0863374113}
|
85 |
+
{"layer":7,"head":0,"avg_attn_entropy":2.1876871879}
|
86 |
+
{"layer":7,"head":1,"avg_attn_entropy":2.7921159844}
|
87 |
+
{"layer":7,"head":2,"avg_attn_entropy":0.4647133289}
|
88 |
+
{"layer":7,"head":3,"avg_attn_entropy":3.0853987454}
|
89 |
+
{"layer":7,"head":4,"avg_attn_entropy":2.8920613989}
|
90 |
+
{"layer":7,"head":5,"avg_attn_entropy":3.1455830612}
|
91 |
+
{"layer":7,"head":6,"avg_attn_entropy":3.07788794}
|
92 |
+
{"layer":7,"head":7,"avg_attn_entropy":3.1755237373}
|
93 |
+
{"layer":7,"head":8,"avg_attn_entropy":0.417414071}
|
94 |
+
{"layer":7,"head":9,"avg_attn_entropy":3.1276816571}
|
95 |
+
{"layer":7,"head":10,"avg_attn_entropy":0.4267832816}
|
96 |
+
{"layer":7,"head":11,"avg_attn_entropy":2.43474786}
|
97 |
+
{"layer":8,"head":0,"avg_attn_entropy":2.6752688019}
|
98 |
+
{"layer":8,"head":1,"avg_attn_entropy":3.5060474741}
|
99 |
+
{"layer":8,"head":2,"avg_attn_entropy":3.3783551812}
|
100 |
+
{"layer":8,"head":3,"avg_attn_entropy":2.8784471669}
|
101 |
+
{"layer":8,"head":4,"avg_attn_entropy":2.8101862297}
|
102 |
+
{"layer":8,"head":5,"avg_attn_entropy":3.1809190528}
|
103 |
+
{"layer":8,"head":6,"avg_attn_entropy":1.2213313031}
|
104 |
+
{"layer":8,"head":7,"avg_attn_entropy":3.0442103526}
|
105 |
+
{"layer":8,"head":8,"avg_attn_entropy":1.3364321963}
|
106 |
+
{"layer":8,"head":9,"avg_attn_entropy":3.0918616837}
|
107 |
+
{"layer":8,"head":10,"avg_attn_entropy":2.9728929253}
|
108 |
+
{"layer":8,"head":11,"avg_attn_entropy":3.0177656777}
|
109 |
+
{"layer":9,"head":0,"avg_attn_entropy":3.1732407826}
|
110 |
+
{"layer":9,"head":1,"avg_attn_entropy":2.6730005873}
|
111 |
+
{"layer":9,"head":2,"avg_attn_entropy":4.1258620846}
|
112 |
+
{"layer":9,"head":3,"avg_attn_entropy":3.2739209476}
|
113 |
+
{"layer":9,"head":4,"avg_attn_entropy":3.2835503089}
|
114 |
+
{"layer":9,"head":5,"avg_attn_entropy":3.4432894998}
|
115 |
+
{"layer":9,"head":6,"avg_attn_entropy":3.728337189}
|
116 |
+
{"layer":9,"head":7,"avg_attn_entropy":3.4111163449}
|
117 |
+
{"layer":9,"head":8,"avg_attn_entropy":3.0758999811}
|
118 |
+
{"layer":9,"head":9,"avg_attn_entropy":3.1317004685}
|
119 |
+
{"layer":9,"head":10,"avg_attn_entropy":3.4239900842}
|
120 |
+
{"layer":9,"head":11,"avg_attn_entropy":2.9890899521}
|
121 |
+
{"layer":10,"head":0,"avg_attn_entropy":3.2778056926}
|
122 |
+
{"layer":10,"head":1,"avg_attn_entropy":3.5493875411}
|
123 |
+
{"layer":10,"head":2,"avg_attn_entropy":3.4417436135}
|
124 |
+
{"layer":10,"head":3,"avg_attn_entropy":3.4047596103}
|
125 |
+
{"layer":10,"head":4,"avg_attn_entropy":3.5894835367}
|
126 |
+
{"layer":10,"head":5,"avg_attn_entropy":3.5689494321}
|
127 |
+
{"layer":10,"head":6,"avg_attn_entropy":3.5860033743}
|
128 |
+
{"layer":10,"head":7,"avg_attn_entropy":3.3755759025}
|
129 |
+
{"layer":10,"head":8,"avg_attn_entropy":3.4994886663}
|
130 |
+
{"layer":10,"head":9,"avg_attn_entropy":3.1777897108}
|
131 |
+
{"layer":10,"head":10,"avg_attn_entropy":3.6837898815}
|
132 |
+
{"layer":10,"head":11,"avg_attn_entropy":2.7966157669}
|
133 |
+
{"layer":11,"head":0,"avg_attn_entropy":3.7790570632}
|
134 |
+
{"layer":11,"head":1,"avg_attn_entropy":3.9694713732}
|
135 |
+
{"layer":11,"head":2,"avg_attn_entropy":3.4806960986}
|
136 |
+
{"layer":11,"head":3,"avg_attn_entropy":3.4770453784}
|
137 |
+
{"layer":11,"head":4,"avg_attn_entropy":3.4994171555}
|
138 |
+
{"layer":11,"head":5,"avg_attn_entropy":3.6039628825}
|
139 |
+
{"layer":11,"head":6,"avg_attn_entropy":2.7959280701}
|
140 |
+
{"layer":11,"head":7,"avg_attn_entropy":3.3850154912}
|
141 |
+
{"layer":11,"head":8,"avg_attn_entropy":3.8393391777}
|
142 |
+
{"layer":11,"head":9,"avg_attn_entropy":3.2341513075}
|
143 |
+
{"layer":11,"head":10,"avg_attn_entropy":3.5350257871}
|
144 |
+
{"layer":11,"head":11,"avg_attn_entropy":3.7378697585}
|