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

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

Neuron 463 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: 0.0537. Min Range: -0.0537

Max Act: 0.0537. Min Act: -0.0006

Data Index: 1310020 (C4 (Web Text))

Max Activating Token Index: 80

Text #1

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0463. Min Act: -0.0005

Data Index: 66631 (C4 (Web Text))

Max Activating Token Index: 782

Text #2

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0456. Min Act: -0.0005

Data Index: 584675 (C4 (Web Text))

Max Activating Token Index: 864

Text #3

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0452. Min Act: -0.0006

Data Index: 991636 (C4 (Web Text))

Max Activating Token Index: 50

Text #4

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0469. Min Act: -0.0006

Data Index: 218362 (C4 (Web Text))

Max Activating Token Index: 262

Text #5

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0424. Min Act: -0.0006

Data Index: 440239 (C4 (Web Text))

Max Activating Token Index: 289

Text #6

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0423. Min Act: -0.0005

Data Index: 564055 (C4 (Web Text))

Max Activating Token Index: 632

Text #7

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0426. Min Act: -0.0005

Data Index: 625333 (C4 (Web Text))

Max Activating Token Index: 128

Text #8

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0429. Min Act: -0.0006

Data Index: 990514 (C4 (Web Text))

Max Activating Token Index: 19

Text #9

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0448. Min Act: -0.0006

Data Index: 1135920 (C4 (Web Text))

Max Activating Token Index: 756

Text #10

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0406. Min Act: -0.0006

Data Index: 1172779 (C4 (Web Text))

Max Activating Token Index: 940

Text #11

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0412. Min Act: -0.0006

Data Index: 1141014 (C4 (Web Text))

Max Activating Token Index: 716

Text #12

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0377. Min Act: -0.0005

Data Index: 1098880 (C4 (Web Text))

Max Activating Token Index: 957

Text #13

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0429. Min Act: -0.0006

Data Index: 431845 (C4 (Web Text))

Max Activating Token Index: 962

Text #14

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0406. Min Act: -0.0006

Data Index: 1512488 (Python Code)

Max Activating Token Index: 232

Text #15

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0397. Min Act: -0.0006

Data Index: 1348806 (C4 (Web Text))

Max Activating Token Index: 957

Text #16

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0412. Min Act: -0.0006

Data Index: 1156520 (C4 (Web Text))

Max Activating Token Index: 717

Text #17

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0394. Min Act: -0.0006

Data Index: 132736 (C4 (Web Text))

Max Activating Token Index: 580

Text #18

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0417. Min Act: -0.0005

Data Index: 1125866 (C4 (Web Text))

Max Activating Token Index: 707

Text #19

Max Range: 0.0537. Min Range: -0.0537

Max Act: 0.0399. Min Act: -0.0005

Data Index: 376515 (C4 (Web Text))

Max Activating Token Index: 551