Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2032 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.1004. Min Act: -0.0006

Data Index: 1620448 (Python Code)

Max Activating Token Index: 159

Text #1

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0799. Min Act: -0.0006

Data Index: 650453 (C4 (Web Text))

Max Activating Token Index: 618

Text #2

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0649. Min Act: -0.0006

Data Index: 1610130 (Python Code)

Max Activating Token Index: 1019

Text #3

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0458. Min Act: -0.0006

Data Index: 150478 (C4 (Web Text))

Max Activating Token Index: 863

Text #4

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0466. Min Act: -0.0006

Data Index: 1504280 (Python Code)

Max Activating Token Index: 85

Text #5

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0448. Min Act: -0.0006

Data Index: 1385715 (Python Code)

Max Activating Token Index: 222

Text #6

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0420. Min Act: -0.0006

Data Index: 1222561 (C4 (Web Text))

Max Activating Token Index: 642

Text #7

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0400. Min Act: -0.0005

Data Index: 1017404 (C4 (Web Text))

Max Activating Token Index: 642

Text #8

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0280. Min Act: -0.0006

Data Index: 27370 (C4 (Web Text))

Max Activating Token Index: 842

Text #9

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0376. Min Act: -0.0006

Data Index: 1059887 (C4 (Web Text))

Max Activating Token Index: 679

Text #10

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0362. Min Act: -0.0006

Data Index: 1368412 (Python Code)

Max Activating Token Index: 968

Text #11

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0379. Min Act: -0.0006

Data Index: 93509 (C4 (Web Text))

Max Activating Token Index: 881

Text #12

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0374. Min Act: -0.0006

Data Index: 288167 (C4 (Web Text))

Max Activating Token Index: 482

Text #13

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0363. Min Act: -0.0006

Data Index: 1387097 (Python Code)

Max Activating Token Index: 870

Text #14

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0365. Min Act: -0.0006

Data Index: 396211 (C4 (Web Text))

Max Activating Token Index: 965

Text #15

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0346. Min Act: -0.0005

Data Index: 460460 (C4 (Web Text))

Max Activating Token Index: 897

Text #16

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0356. Min Act: -0.0006

Data Index: 530701 (C4 (Web Text))

Max Activating Token Index: 831

Text #17

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0348. Min Act: -0.0005

Data Index: 935537 (C4 (Web Text))

Max Activating Token Index: 956

Text #18

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0344. Min Act: -0.0006

Data Index: 1174476 (C4 (Web Text))

Max Activating Token Index: 1008

Text #19

Max Range: 0.1004. Min Range: -0.1004

Max Act: 0.0326. Min Act: -0.0006

Data Index: 621703 (C4 (Web Text))

Max Activating Token Index: 815