Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3134 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.9225. Min Act: -0.0000

Data Index: 1595925 (Python Code)

Max Activating Token Index: 719

Text #1

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.7639. Min Act: -0.0000

Data Index: 687188 (C4 (Web Text))

Max Activating Token Index: 167

Text #2

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.7340. Min Act: -0.0000

Data Index: 1551112 (Python Code)

Max Activating Token Index: 1009

Text #3

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.6639. Min Act: -0.0000

Data Index: 1394779 (Python Code)

Max Activating Token Index: 879

Text #4

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.6052. Min Act: -0.0000

Data Index: 209624 (C4 (Web Text))

Max Activating Token Index: 350

Text #5

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.5766. Min Act: -0.0000

Data Index: 938770 (C4 (Web Text))

Max Activating Token Index: 506

Text #6

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.5534. Min Act: -0.0000

Data Index: 1069949 (C4 (Web Text))

Max Activating Token Index: 140

Text #7

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.1895. Min Act: -0.0000

Data Index: 1454048 (Python Code)

Max Activating Token Index: 606

Text #8

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.3826. Min Act: -0.0000

Data Index: 1215703 (C4 (Web Text))

Max Activating Token Index: 195

Text #9

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.3832. Min Act: -0.0000

Data Index: 509352 (C4 (Web Text))

Max Activating Token Index: 277

Text #10

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.4828. Min Act: -0.0000

Data Index: 1627508 (Python Code)

Max Activating Token Index: 591

Text #11

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.4498. Min Act: -0.0000

Data Index: 1068760 (C4 (Web Text))

Max Activating Token Index: 216

Text #12

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.2054. Min Act: -0.0000

Data Index: 1401102 (Python Code)

Max Activating Token Index: 327

Text #13

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.4232. Min Act: -0.0000

Data Index: 954822 (C4 (Web Text))

Max Activating Token Index: 642

Text #14

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.4127. Min Act: -0.0000

Data Index: 1475212 (Python Code)

Max Activating Token Index: 235

Text #15

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.4301. Min Act: -0.0000

Data Index: 1360543 (Python Code)

Max Activating Token Index: 614

Text #16

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.1357. Min Act: -0.0000

Data Index: 1597884 (Python Code)

Max Activating Token Index: 983

Text #17

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.3794. Min Act: -0.0000

Data Index: 301760 (C4 (Web Text))

Max Activating Token Index: 184

Text #18

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.2456. Min Act: -0.0000

Data Index: 1448098 (Python Code)

Max Activating Token Index: 934

Text #19

Max Range: 2.9225. Min Range: -2.9225

Max Act: 2.3716. Min Act: -0.0000

Data Index: 290255 (C4 (Web Text))

Max Activating Token Index: 922