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":5.3037984309}
|
2 |
+
{"layer":0,"head":1,"avg_attn_entropy":5.4211173139}
|
3 |
+
{"layer":0,"head":2,"avg_attn_entropy":5.6192001915}
|
4 |
+
{"layer":0,"head":3,"avg_attn_entropy":5.5837948513}
|
5 |
+
{"layer":0,"head":4,"avg_attn_entropy":5.5580151672}
|
6 |
+
{"layer":0,"head":5,"avg_attn_entropy":5.8510678463}
|
7 |
+
{"layer":0,"head":6,"avg_attn_entropy":5.7067425313}
|
8 |
+
{"layer":0,"head":7,"avg_attn_entropy":5.5785931864}
|
9 |
+
{"layer":0,"head":8,"avg_attn_entropy":5.5048454089}
|
10 |
+
{"layer":0,"head":9,"avg_attn_entropy":4.9825316315}
|
11 |
+
{"layer":0,"head":10,"avg_attn_entropy":5.4533094587}
|
12 |
+
{"layer":0,"head":11,"avg_attn_entropy":4.404718461}
|
13 |
+
{"layer":1,"head":0,"avg_attn_entropy":4.3423261659}
|
14 |
+
{"layer":1,"head":1,"avg_attn_entropy":4.3697919683}
|
15 |
+
{"layer":1,"head":2,"avg_attn_entropy":5.4056465788}
|
16 |
+
{"layer":1,"head":3,"avg_attn_entropy":3.2884810832}
|
17 |
+
{"layer":1,"head":4,"avg_attn_entropy":4.882887351}
|
18 |
+
{"layer":1,"head":5,"avg_attn_entropy":4.4652225971}
|
19 |
+
{"layer":1,"head":6,"avg_attn_entropy":2.252860395}
|
20 |
+
{"layer":1,"head":7,"avg_attn_entropy":0.0473486412}
|
21 |
+
{"layer":1,"head":8,"avg_attn_entropy":5.6761930828}
|
22 |
+
{"layer":1,"head":9,"avg_attn_entropy":5.3927624536}
|
23 |
+
{"layer":1,"head":10,"avg_attn_entropy":4.2997572675}
|
24 |
+
{"layer":1,"head":11,"avg_attn_entropy":5.4956344876}
|
25 |
+
{"layer":2,"head":0,"avg_attn_entropy":0.0246000381}
|
26 |
+
{"layer":2,"head":1,"avg_attn_entropy":5.3775267115}
|
27 |
+
{"layer":2,"head":2,"avg_attn_entropy":4.5741401463}
|
28 |
+
{"layer":2,"head":3,"avg_attn_entropy":5.3182373524}
|
29 |
+
{"layer":2,"head":4,"avg_attn_entropy":4.2277917414}
|
30 |
+
{"layer":2,"head":5,"avg_attn_entropy":5.1080817356}
|
31 |
+
{"layer":2,"head":6,"avg_attn_entropy":2.0771291815}
|
32 |
+
{"layer":2,"head":7,"avg_attn_entropy":5.3094343677}
|
33 |
+
{"layer":2,"head":8,"avg_attn_entropy":5.2134828053}
|
34 |
+
{"layer":2,"head":9,"avg_attn_entropy":4.8007494488}
|
35 |
+
{"layer":2,"head":10,"avg_attn_entropy":0.5755346746}
|
36 |
+
{"layer":2,"head":11,"avg_attn_entropy":2.3393019724}
|
37 |
+
{"layer":3,"head":0,"avg_attn_entropy":4.7174438205}
|
38 |
+
{"layer":3,"head":1,"avg_attn_entropy":3.3079878941}
|
39 |
+
{"layer":3,"head":2,"avg_attn_entropy":4.0613764279}
|
40 |
+
{"layer":3,"head":3,"avg_attn_entropy":4.8474606782}
|
41 |
+
{"layer":3,"head":4,"avg_attn_entropy":5.1030845609}
|
42 |
+
{"layer":3,"head":5,"avg_attn_entropy":4.1798858626}
|
43 |
+
{"layer":3,"head":6,"avg_attn_entropy":4.7629247117}
|
44 |
+
{"layer":3,"head":7,"avg_attn_entropy":4.9996004438}
|
45 |
+
{"layer":3,"head":8,"avg_attn_entropy":4.2472870538}
|
46 |
+
{"layer":3,"head":9,"avg_attn_entropy":4.4079606977}
|
47 |
+
{"layer":3,"head":10,"avg_attn_entropy":4.45142693}
|
48 |
+
{"layer":3,"head":11,"avg_attn_entropy":4.8568112419}
|
49 |
+
{"layer":4,"head":0,"avg_attn_entropy":5.1378735404}
|
50 |
+
{"layer":4,"head":1,"avg_attn_entropy":5.1280567396}
|
51 |
+
{"layer":4,"head":2,"avg_attn_entropy":5.3690518191}
|
52 |
+
{"layer":4,"head":3,"avg_attn_entropy":4.845268213}
|
53 |
+
{"layer":4,"head":4,"avg_attn_entropy":2.9069557767}
|
54 |
+
{"layer":4,"head":5,"avg_attn_entropy":4.0366454403}
|
55 |
+
{"layer":4,"head":6,"avg_attn_entropy":4.26148296}
|
56 |
+
{"layer":4,"head":7,"avg_attn_entropy":5.2084208949}
|
57 |
+
{"layer":4,"head":8,"avg_attn_entropy":4.2795615587}
|
58 |
+
{"layer":4,"head":9,"avg_attn_entropy":4.8731250048}
|
59 |
+
{"layer":4,"head":10,"avg_attn_entropy":3.2074705315}
|
60 |
+
{"layer":4,"head":11,"avg_attn_entropy":2.52117693}
|
61 |
+
{"layer":5,"head":0,"avg_attn_entropy":4.1590600086}
|
62 |
+
{"layer":5,"head":1,"avg_attn_entropy":4.3579832439}
|
63 |
+
{"layer":5,"head":2,"avg_attn_entropy":3.3265648817}
|
64 |
+
{"layer":5,"head":3,"avg_attn_entropy":3.9188816745}
|
65 |
+
{"layer":5,"head":4,"avg_attn_entropy":3.2287086642}
|
66 |
+
{"layer":5,"head":5,"avg_attn_entropy":0.1971491879}
|
67 |
+
{"layer":5,"head":6,"avg_attn_entropy":2.9136548887}
|
68 |
+
{"layer":5,"head":7,"avg_attn_entropy":4.1633383267}
|
69 |
+
{"layer":5,"head":8,"avg_attn_entropy":2.6176570553}
|
70 |
+
{"layer":5,"head":9,"avg_attn_entropy":3.8337873371}
|
71 |
+
{"layer":5,"head":10,"avg_attn_entropy":2.8518897671}
|
72 |
+
{"layer":5,"head":11,"avg_attn_entropy":3.1321802158}
|
73 |
+
{"layer":6,"head":0,"avg_attn_entropy":3.1499839302}
|
74 |
+
{"layer":6,"head":1,"avg_attn_entropy":2.6422339039}
|
75 |
+
{"layer":6,"head":2,"avg_attn_entropy":3.6701242104}
|
76 |
+
{"layer":6,"head":3,"avg_attn_entropy":3.4167328811}
|
77 |
+
{"layer":6,"head":4,"avg_attn_entropy":3.4019336029}
|
78 |
+
{"layer":6,"head":5,"avg_attn_entropy":3.7282611787}
|
79 |
+
{"layer":6,"head":6,"avg_attn_entropy":3.512815958}
|
80 |
+
{"layer":6,"head":7,"avg_attn_entropy":3.4759070677}
|
81 |
+
{"layer":6,"head":8,"avg_attn_entropy":3.768742987}
|
82 |
+
{"layer":6,"head":9,"avg_attn_entropy":5.0586992462}
|
83 |
+
{"layer":6,"head":10,"avg_attn_entropy":3.8829400187}
|
84 |
+
{"layer":6,"head":11,"avg_attn_entropy":2.0566005801}
|
85 |
+
{"layer":7,"head":0,"avg_attn_entropy":2.2384937618}
|
86 |
+
{"layer":7,"head":1,"avg_attn_entropy":2.7942546206}
|
87 |
+
{"layer":7,"head":2,"avg_attn_entropy":0.5244402446}
|
88 |
+
{"layer":7,"head":3,"avg_attn_entropy":3.0675714668}
|
89 |
+
{"layer":7,"head":4,"avg_attn_entropy":2.9298430017}
|
90 |
+
{"layer":7,"head":5,"avg_attn_entropy":3.1482338706}
|
91 |
+
{"layer":7,"head":6,"avg_attn_entropy":3.0806305526}
|
92 |
+
{"layer":7,"head":7,"avg_attn_entropy":3.1991121953}
|
93 |
+
{"layer":7,"head":8,"avg_attn_entropy":0.4262889694}
|
94 |
+
{"layer":7,"head":9,"avg_attn_entropy":3.1550252758}
|
95 |
+
{"layer":7,"head":10,"avg_attn_entropy":0.4899320335}
|
96 |
+
{"layer":7,"head":11,"avg_attn_entropy":2.5529023086}
|
97 |
+
{"layer":8,"head":0,"avg_attn_entropy":2.7152027614}
|
98 |
+
{"layer":8,"head":1,"avg_attn_entropy":3.5020538679}
|
99 |
+
{"layer":8,"head":2,"avg_attn_entropy":3.3980820777}
|
100 |
+
{"layer":8,"head":3,"avg_attn_entropy":2.9082141902}
|
101 |
+
{"layer":8,"head":4,"avg_attn_entropy":2.8977757412}
|
102 |
+
{"layer":8,"head":5,"avg_attn_entropy":3.2504383882}
|
103 |
+
{"layer":8,"head":6,"avg_attn_entropy":1.3441756091}
|
104 |
+
{"layer":8,"head":7,"avg_attn_entropy":3.0555697059}
|
105 |
+
{"layer":8,"head":8,"avg_attn_entropy":1.3895034982}
|
106 |
+
{"layer":8,"head":9,"avg_attn_entropy":3.1127950507}
|
107 |
+
{"layer":8,"head":10,"avg_attn_entropy":2.9625080585}
|
108 |
+
{"layer":8,"head":11,"avg_attn_entropy":3.0436740361}
|
109 |
+
{"layer":9,"head":0,"avg_attn_entropy":3.2192375689}
|
110 |
+
{"layer":9,"head":1,"avg_attn_entropy":2.7753029476}
|
111 |
+
{"layer":9,"head":2,"avg_attn_entropy":4.1071542175}
|
112 |
+
{"layer":9,"head":3,"avg_attn_entropy":3.314338016}
|
113 |
+
{"layer":9,"head":4,"avg_attn_entropy":3.316797262}
|
114 |
+
{"layer":9,"head":5,"avg_attn_entropy":3.4764415892}
|
115 |
+
{"layer":9,"head":6,"avg_attn_entropy":3.6949476767}
|
116 |
+
{"layer":9,"head":7,"avg_attn_entropy":3.4307457755}
|
117 |
+
{"layer":9,"head":8,"avg_attn_entropy":3.0914603593}
|
118 |
+
{"layer":9,"head":9,"avg_attn_entropy":3.1517226637}
|
119 |
+
{"layer":9,"head":10,"avg_attn_entropy":3.4211373632}
|
120 |
+
{"layer":9,"head":11,"avg_attn_entropy":3.0040744163}
|
121 |
+
{"layer":10,"head":0,"avg_attn_entropy":3.2979680163}
|
122 |
+
{"layer":10,"head":1,"avg_attn_entropy":3.5520227871}
|
123 |
+
{"layer":10,"head":2,"avg_attn_entropy":3.453571879}
|
124 |
+
{"layer":10,"head":3,"avg_attn_entropy":3.3913325174}
|
125 |
+
{"layer":10,"head":4,"avg_attn_entropy":3.585930636}
|
126 |
+
{"layer":10,"head":5,"avg_attn_entropy":3.5685538169}
|
127 |
+
{"layer":10,"head":6,"avg_attn_entropy":3.5539147525}
|
128 |
+
{"layer":10,"head":7,"avg_attn_entropy":3.3957042848}
|
129 |
+
{"layer":10,"head":8,"avg_attn_entropy":3.5230200355}
|
130 |
+
{"layer":10,"head":9,"avg_attn_entropy":3.1195353184}
|
131 |
+
{"layer":10,"head":10,"avg_attn_entropy":3.6956416023}
|
132 |
+
{"layer":10,"head":11,"avg_attn_entropy":2.8039052529}
|
133 |
+
{"layer":11,"head":0,"avg_attn_entropy":3.7792830617}
|
134 |
+
{"layer":11,"head":1,"avg_attn_entropy":4.0092900038}
|
135 |
+
{"layer":11,"head":2,"avg_attn_entropy":3.4837087975}
|
136 |
+
{"layer":11,"head":3,"avg_attn_entropy":3.471751188}
|
137 |
+
{"layer":11,"head":4,"avg_attn_entropy":3.4896705894}
|
138 |
+
{"layer":11,"head":5,"avg_attn_entropy":3.5876569436}
|
139 |
+
{"layer":11,"head":6,"avg_attn_entropy":2.7622260865}
|
140 |
+
{"layer":11,"head":7,"avg_attn_entropy":3.357837483}
|
141 |
+
{"layer":11,"head":8,"avg_attn_entropy":3.8535885484}
|
142 |
+
{"layer":11,"head":9,"avg_attn_entropy":3.1898954339}
|
143 |
+
{"layer":11,"head":10,"avg_attn_entropy":3.5580659375}
|
144 |
+
{"layer":11,"head":11,"avg_attn_entropy":3.7243528434}
|