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

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

Neuron 2984 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3229. Min Range: -1.3229

Max Act: 1.3229. Min Act: -0.0000

Data Index: 489191 (C4 (Web Text))

Max Activating Token Index: 753

Text #1

Max Range: 1.3229. Min Range: -1.3229

Max Act: 1.0687. Min Act: -0.0000

Data Index: 1617329 (Python Code)

Max Activating Token Index: 50

Text #2

Max Range: 1.3229. Min Range: -1.3229

Max Act: 1.1459. Min Act: -0.0000

Data Index: 1208402 (C4 (Web Text))

Max Activating Token Index: 426

Text #3

Max Range: 1.3229. Min Range: -1.3229

Max Act: 1.1176. Min Act: -0.0000

Data Index: 1527827 (Python Code)

Max Activating Token Index: 432

Text #4

Max Range: 1.3229. Min Range: -1.3229

Max Act: 1.0052. Min Act: -0.0000

Data Index: 549234 (C4 (Web Text))

Max Activating Token Index: 515

Text #5

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.9012. Min Act: -0.0000

Data Index: 968098 (C4 (Web Text))

Max Activating Token Index: 296

Text #6

Max Range: 1.3229. Min Range: -1.3229

Max Act: 1.0318. Min Act: -0.0000

Data Index: 1614632 (Python Code)

Max Activating Token Index: 973

Text #7

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.9227. Min Act: -0.0000

Data Index: 1446510 (Python Code)

Max Activating Token Index: 206

Text #8

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.9919. Min Act: -0.0000

Data Index: 418466 (C4 (Web Text))

Max Activating Token Index: 595

Text #9

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.9497. Min Act: -0.0000

Data Index: 1149549 (C4 (Web Text))

Max Activating Token Index: 34

Text #10

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.9482. Min Act: -0.0000

Data Index: 1605539 (Python Code)

Max Activating Token Index: 676

Text #11

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.7771. Min Act: -0.0000

Data Index: 1477080 (Python Code)

Max Activating Token Index: 823

Text #12

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.8520. Min Act: -0.0000

Data Index: 869087 (C4 (Web Text))

Max Activating Token Index: 777

Text #13

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.8003. Min Act: -0.0000

Data Index: 373943 (C4 (Web Text))

Max Activating Token Index: 289

Text #14

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.8135. Min Act: -0.0000

Data Index: 219701 (C4 (Web Text))

Max Activating Token Index: 976

Text #15

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.8632. Min Act: -0.0000

Data Index: 1145918 (C4 (Web Text))

Max Activating Token Index: 451

Text #16

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.8031. Min Act: -0.0000

Data Index: 1053676 (C4 (Web Text))

Max Activating Token Index: 89

Text #17

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.9334. Min Act: -0.0000

Data Index: 1538188 (Python Code)

Max Activating Token Index: 813

Text #18

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.7955. Min Act: -0.0000

Data Index: 1648403 (Python Code)

Max Activating Token Index: 56

Text #19

Max Range: 1.3229. Min Range: -1.3229

Max Act: 0.8236. Min Act: -0.0000

Data Index: 16158 (C4 (Web Text))

Max Activating Token Index: 570