Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1179 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0547. Min Act: -0.0002

Data Index: 680331 (C4 (Web Text))

Max Activating Token Index: 543

Text #1

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0458. Min Act: -0.0002

Data Index: 62523 (C4 (Web Text))

Max Activating Token Index: 278

Text #2

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0439. Min Act: -0.0002

Data Index: 426253 (C4 (Web Text))

Max Activating Token Index: 464

Text #3

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0429. Min Act: -0.0002

Data Index: 557566 (C4 (Web Text))

Max Activating Token Index: 503

Text #4

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0443. Min Act: -0.0002

Data Index: 1097367 (C4 (Web Text))

Max Activating Token Index: 835

Text #5

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0441. Min Act: -0.0002

Data Index: 853441 (C4 (Web Text))

Max Activating Token Index: 791

Text #6

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0446. Min Act: -0.0001

Data Index: 268271 (C4 (Web Text))

Max Activating Token Index: 575

Text #7

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0420. Min Act: -0.0002

Data Index: 41793 (C4 (Web Text))

Max Activating Token Index: 614

Text #8

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0408. Min Act: -0.0002

Data Index: 1149084 (C4 (Web Text))

Max Activating Token Index: 908

Text #9

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0399. Min Act: -0.0002

Data Index: 1286017 (C4 (Web Text))

Max Activating Token Index: 135

Text #10

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0401. Min Act: -0.0001

Data Index: 1109912 (C4 (Web Text))

Max Activating Token Index: 574

Text #11

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0356. Min Act: -0.0002

Data Index: 1589125 (Python Code)

Max Activating Token Index: 887

Text #12

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0392. Min Act: -0.0002

Data Index: 1075044 (C4 (Web Text))

Max Activating Token Index: 127

Text #13

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0378. Min Act: -0.0002

Data Index: 1519861 (Python Code)

Max Activating Token Index: 823

Text #14

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0377. Min Act: -0.0002

Data Index: 387323 (C4 (Web Text))

Max Activating Token Index: 389

Text #15

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0398. Min Act: -0.0002

Data Index: 815990 (C4 (Web Text))

Max Activating Token Index: 381

Text #16

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0386. Min Act: -0.0001

Data Index: 1141185 (C4 (Web Text))

Max Activating Token Index: 657

Text #17

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0399. Min Act: -0.0002

Data Index: 423564 (C4 (Web Text))

Max Activating Token Index: 124

Text #18

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0364. Min Act: -0.0002

Data Index: 524754 (C4 (Web Text))

Max Activating Token Index: 425

Text #19

Max Range: 0.0547. Min Range: -0.0547

Max Act: 0.0388. Min Act: -0.0002

Data Index: 665138 (C4 (Web Text))

Max Activating Token Index: 19