Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 975 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0527. Min Act: -0.0001

Data Index: 1232405 (C4 (Web Text))

Max Activating Token Index: 21

Text #1

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0482. Min Act: -0.0001

Data Index: 1463691 (Python Code)

Max Activating Token Index: 43

Text #2

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0440. Min Act: -0.0001

Data Index: 63717 (C4 (Web Text))

Max Activating Token Index: 36

Text #3

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0456. Min Act: -0.0001

Data Index: 1041395 (C4 (Web Text))

Max Activating Token Index: 123

Text #4

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0444. Min Act: -0.0001

Data Index: 580778 (C4 (Web Text))

Max Activating Token Index: 375

Text #5

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0435. Min Act: -0.0001

Data Index: 469970 (C4 (Web Text))

Max Activating Token Index: 13

Text #6

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0414. Min Act: -0.0001

Data Index: 1627084 (Python Code)

Max Activating Token Index: 25

Text #7

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0409. Min Act: -0.0001

Data Index: 1166593 (C4 (Web Text))

Max Activating Token Index: 128

Text #8

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0419. Min Act: -0.0001

Data Index: 1297999 (C4 (Web Text))

Max Activating Token Index: 824

Text #9

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0419. Min Act: -0.0001

Data Index: 223995 (C4 (Web Text))

Max Activating Token Index: 484

Text #10

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0385. Min Act: -0.0001

Data Index: 1524827 (Python Code)

Max Activating Token Index: 716

Text #11

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0385. Min Act: -0.0001

Data Index: 1611992 (Python Code)

Max Activating Token Index: 73

Text #12

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0400. Min Act: -0.0001

Data Index: 636731 (C4 (Web Text))

Max Activating Token Index: 787

Text #13

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0397. Min Act: -0.0001

Data Index: 1564859 (Python Code)

Max Activating Token Index: 68

Text #14

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0381. Min Act: -0.0001

Data Index: 317383 (C4 (Web Text))

Max Activating Token Index: 879

Text #15

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0397. Min Act: -0.0001

Data Index: 1654531 (Python Code)

Max Activating Token Index: 39

Text #16

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0393. Min Act: -0.0001

Data Index: 570904 (C4 (Web Text))

Max Activating Token Index: 18

Text #17

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0380. Min Act: -0.0001

Data Index: 867754 (C4 (Web Text))

Max Activating Token Index: 16

Text #18

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0377. Min Act: -0.0001

Data Index: 1502650 (Python Code)

Max Activating Token Index: 25

Text #19

Max Range: 0.0527. Min Range: -0.0527

Max Act: 0.0377. Min Act: -0.0001

Data Index: 1005040 (C4 (Web Text))

Max Activating Token Index: 346