Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 1667 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1252. Min Act: -0.0004

Data Index: 993023 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1125. Min Act: -0.0003

Data Index: 1528383 (Python Code)

Max Activating Token Index: 672

Text #2

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1151. Min Act: -0.0003

Data Index: 1373439 (Python Code)

Max Activating Token Index: 520

Text #3

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1148. Min Act: -0.0003

Data Index: 863150 (C4 (Web Text))

Max Activating Token Index: 417

Text #4

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1109. Min Act: -0.0003

Data Index: 1289622 (C4 (Web Text))

Max Activating Token Index: 634

Text #5

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1113. Min Act: -0.0003

Data Index: 1446804 (Python Code)

Max Activating Token Index: 900

Text #6

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1098. Min Act: -0.0003

Data Index: 1035152 (C4 (Web Text))

Max Activating Token Index: 131

Text #7

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1082. Min Act: -0.0003

Data Index: 1305402 (C4 (Web Text))

Max Activating Token Index: 214

Text #8

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1067. Min Act: -0.0004

Data Index: 467922 (C4 (Web Text))

Max Activating Token Index: 303

Text #9

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1061. Min Act: -0.0004

Data Index: 476073 (C4 (Web Text))

Max Activating Token Index: 556

Text #10

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1060. Min Act: -0.0003

Data Index: 646717 (C4 (Web Text))

Max Activating Token Index: 663

Text #11

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1075. Min Act: -0.0003

Data Index: 36138 (C4 (Web Text))

Max Activating Token Index: 725

Text #12

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1041. Min Act: -0.0003

Data Index: 562761 (C4 (Web Text))

Max Activating Token Index: 579

Text #13

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1078. Min Act: -0.0003

Data Index: 1427727 (Python Code)

Max Activating Token Index: 625

Text #14

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1030. Min Act: -0.0003

Data Index: 1098260 (C4 (Web Text))

Max Activating Token Index: 284

Text #15

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1004. Min Act: -0.0004

Data Index: 214195 (C4 (Web Text))

Max Activating Token Index: 445

Text #16

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1031. Min Act: -0.0003

Data Index: 506051 (C4 (Web Text))

Max Activating Token Index: 344

Text #17

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1039. Min Act: -0.0003

Data Index: 511742 (C4 (Web Text))

Max Activating Token Index: 302

Text #18

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1032. Min Act: -0.0003

Data Index: 821868 (C4 (Web Text))

Max Activating Token Index: 55

Text #19

Max Range: 0.1252. Min Range: -0.1252

Max Act: 0.1066. Min Act: -0.0003

Data Index: 619260 (C4 (Web Text))

Max Activating Token Index: 478