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

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

Neuron 4503 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.8384. Min Act: -0.0000

Data Index: 772199 (C4 (Web Text))

Max Activating Token Index: 667

Text #1

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.3965. Min Act: -0.0000

Data Index: 1283277 (C4 (Web Text))

Max Activating Token Index: 528

Text #2

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.2412. Min Act: -0.0000

Data Index: 921168 (C4 (Web Text))

Max Activating Token Index: 186

Text #3

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.0723. Min Act: -0.0000

Data Index: 955385 (C4 (Web Text))

Max Activating Token Index: 812

Text #4

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.1883. Min Act: -0.0000

Data Index: 1386062 (Python Code)

Max Activating Token Index: 920

Text #5

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.1556. Min Act: -0.0000

Data Index: 1227783 (C4 (Web Text))

Max Activating Token Index: 421

Text #6

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.1212. Min Act: -0.0000

Data Index: 213789 (C4 (Web Text))

Max Activating Token Index: 385

Text #7

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.0172. Min Act: -0.0000

Data Index: 71087 (C4 (Web Text))

Max Activating Token Index: 573

Text #8

Max Range: 2.8384. Min Range: -2.8384

Max Act: 2.1181. Min Act: -0.0000

Data Index: 937283 (C4 (Web Text))

Max Activating Token Index: 81

Text #9

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.8843. Min Act: -0.0000

Data Index: 1320197 (C4 (Web Text))

Max Activating Token Index: 43

Text #10

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.8859. Min Act: -0.0000

Data Index: 329678 (C4 (Web Text))

Max Activating Token Index: 223

Text #11

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.7780. Min Act: -0.0000

Data Index: 398510 (C4 (Web Text))

Max Activating Token Index: 610

Text #12

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.7614. Min Act: -0.0000

Data Index: 436090 (C4 (Web Text))

Max Activating Token Index: 857

Text #13

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.7001. Min Act: -0.0000

Data Index: 1420285 (Python Code)

Max Activating Token Index: 426

Text #14

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.8036. Min Act: -0.0000

Data Index: 899621 (C4 (Web Text))

Max Activating Token Index: 86

Text #15

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.6655. Min Act: -0.0000

Data Index: 750660 (C4 (Web Text))

Max Activating Token Index: 677

Text #16

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.6612. Min Act: -0.0000

Data Index: 158401 (C4 (Web Text))

Max Activating Token Index: 692

Text #17

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.6391. Min Act: -0.0000

Data Index: 378311 (C4 (Web Text))

Max Activating Token Index: 1016

Text #18

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.5195. Min Act: -0.0000

Data Index: 1414096 (Python Code)

Max Activating Token Index: 252

Text #19

Max Range: 2.8384. Min Range: -2.8384

Max Act: 1.5301. Min Act: -0.0000

Data Index: 855490 (C4 (Web Text))

Max Activating Token Index: 898