Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 922 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1307. Min Act: -0.0001

Data Index: 629153 (C4 (Web Text))

Max Activating Token Index: 241

Text #1

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1256. Min Act: -0.0001

Data Index: 37252 (C4 (Web Text))

Max Activating Token Index: 46

Text #2

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1212. Min Act: -0.0001

Data Index: 723609 (C4 (Web Text))

Max Activating Token Index: 830

Text #3

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1210. Min Act: -0.0001

Data Index: 1276594 (C4 (Web Text))

Max Activating Token Index: 817

Text #4

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1188. Min Act: -0.0001

Data Index: 582717 (C4 (Web Text))

Max Activating Token Index: 127

Text #5

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1190. Min Act: -0.0001

Data Index: 800577 (C4 (Web Text))

Max Activating Token Index: 33

Text #6

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1183. Min Act: -0.0001

Data Index: 814040 (C4 (Web Text))

Max Activating Token Index: 336

Text #7

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1156. Min Act: -0.0001

Data Index: 499183 (C4 (Web Text))

Max Activating Token Index: 479

Text #8

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1176. Min Act: -0.0001

Data Index: 1388433 (Python Code)

Max Activating Token Index: 332

Text #9

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1175. Min Act: -0.0001

Data Index: 222415 (C4 (Web Text))

Max Activating Token Index: 300

Text #10

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1170. Min Act: -0.0001

Data Index: 1634845 (Python Code)

Max Activating Token Index: 327

Text #11

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1118. Min Act: -0.0001

Data Index: 1185474 (C4 (Web Text))

Max Activating Token Index: 736

Text #12

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1147. Min Act: -0.0001

Data Index: 1379638 (Python Code)

Max Activating Token Index: 687

Text #13

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1108. Min Act: -0.0001

Data Index: 1019519 (C4 (Web Text))

Max Activating Token Index: 133

Text #14

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1103. Min Act: -0.0001

Data Index: 560019 (C4 (Web Text))

Max Activating Token Index: 333

Text #15

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1084. Min Act: -0.0001

Data Index: 1615338 (Python Code)

Max Activating Token Index: 769

Text #16

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1061. Min Act: -0.0001

Data Index: 666406 (C4 (Web Text))

Max Activating Token Index: 785

Text #17

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1053. Min Act: -0.0001

Data Index: 887879 (C4 (Web Text))

Max Activating Token Index: 472

Text #18

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1083. Min Act: -0.0001

Data Index: 450916 (C4 (Web Text))

Max Activating Token Index: 93

Text #19

Max Range: 0.1307. Min Range: -0.1307

Max Act: 0.1046. Min Act: -0.0001

Data Index: 516889 (C4 (Web Text))

Max Activating Token Index: 847