Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: The Pile

Neuron 4701 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.6459. Min Act: -0.0000

Data Index: 1220164 (The Pile)

Max Activating Token Index: 51

Text #1

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.4891. Min Act: -0.0000

Data Index: 13003 (The Pile)

Max Activating Token Index: 45

Text #2

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.6395. Min Act: -0.0000

Data Index: 1222881 (The Pile)

Max Activating Token Index: 54

Text #3

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.8288. Min Act: -0.0000

Data Index: 1328024 (The Pile)

Max Activating Token Index: 133

Text #4

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.4006. Min Act: -0.0000

Data Index: 1607144 (The Pile)

Max Activating Token Index: 161

Text #5

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.3975. Min Act: -0.0000

Data Index: 1596714 (The Pile)

Max Activating Token Index: 459

Text #6

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.2128. Min Act: -0.0000

Data Index: 1350546 (The Pile)

Max Activating Token Index: 52

Text #7

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.1846. Min Act: -0.0000

Data Index: 920557 (The Pile)

Max Activating Token Index: 92

Text #8

Max Range: 7.8288. Min Range: -7.8288

Max Act: 6.9885. Min Act: -0.0000

Data Index: 1179122 (The Pile)

Max Activating Token Index: 92

Text #9

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.2933. Min Act: -0.0000

Data Index: 1579959 (The Pile)

Max Activating Token Index: 417

Text #10

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.4407. Min Act: -0.0000

Data Index: 1731050 (The Pile)

Max Activating Token Index: 51

Text #11

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.1172. Min Act: -0.0000

Data Index: 1100384 (The Pile)

Max Activating Token Index: 54

Text #12

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.2443. Min Act: -0.0000

Data Index: 1740721 (The Pile)

Max Activating Token Index: 61

Text #13

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.2089. Min Act: -0.0000

Data Index: 1925550 (The Pile)

Max Activating Token Index: 63

Text #14

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.3265. Min Act: -0.0000

Data Index: 1553390 (The Pile)

Max Activating Token Index: 53

Text #15

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.3320. Min Act: -0.0000

Data Index: 1361913 (The Pile)

Max Activating Token Index: 78

Text #16

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.0391. Min Act: -0.0000

Data Index: 15759 (The Pile)

Max Activating Token Index: 428

Text #17

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.2138. Min Act: -0.0000

Data Index: 232318 (The Pile)

Max Activating Token Index: 151

Text #18

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.0136. Min Act: -0.0000

Data Index: 9210 (The Pile)

Max Activating Token Index: 181

Text #19

Max Range: 7.8288. Min Range: -7.8288

Max Act: 7.0103. Min Act: -0.0000

Data Index: 1233812 (The Pile)

Max Activating Token Index: 75