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

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

Neuron 422 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: 0.4117. Min Range: -0.4117

Max Act: 0.4117. Min Act: -0.0001

Data Index: 925091 (C4 (Web Text))

Max Activating Token Index: 135

Text #1

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3999. Min Act: -0.0001

Data Index: 329771 (C4 (Web Text))

Max Activating Token Index: 154

Text #2

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3658. Min Act: -0.0001

Data Index: 395936 (C4 (Web Text))

Max Activating Token Index: 941

Text #3

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3603. Min Act: -0.0001

Data Index: 340399 (C4 (Web Text))

Max Activating Token Index: 79

Text #4

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3458. Min Act: -0.0001

Data Index: 1310944 (C4 (Web Text))

Max Activating Token Index: 516

Text #5

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3546. Min Act: -0.0001

Data Index: 710249 (C4 (Web Text))

Max Activating Token Index: 80

Text #6

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3415. Min Act: -0.0001

Data Index: 636196 (C4 (Web Text))

Max Activating Token Index: 954

Text #7

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3425. Min Act: -0.0001

Data Index: 448582 (C4 (Web Text))

Max Activating Token Index: 36

Text #8

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3501. Min Act: -0.0001

Data Index: 1007854 (C4 (Web Text))

Max Activating Token Index: 958

Text #9

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3646. Min Act: -0.0001

Data Index: 1209768 (C4 (Web Text))

Max Activating Token Index: 894

Text #10

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3408. Min Act: -0.0001

Data Index: 1524299 (Python Code)

Max Activating Token Index: 723

Text #11

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3300. Min Act: -0.0001

Data Index: 1375134 (Python Code)

Max Activating Token Index: 769

Text #12

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3645. Min Act: -0.0001

Data Index: 188194 (C4 (Web Text))

Max Activating Token Index: 801

Text #13

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3380. Min Act: -0.0001

Data Index: 1641308 (Python Code)

Max Activating Token Index: 793

Text #14

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3368. Min Act: -0.0001

Data Index: 1065636 (C4 (Web Text))

Max Activating Token Index: 999

Text #15

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3356. Min Act: -0.0001

Data Index: 65787 (C4 (Web Text))

Max Activating Token Index: 981

Text #16

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3349. Min Act: -0.0001

Data Index: 465098 (C4 (Web Text))

Max Activating Token Index: 822

Text #17

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3565. Min Act: -0.0001

Data Index: 693102 (C4 (Web Text))

Max Activating Token Index: 767

Text #18

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3320. Min Act: -0.0001

Data Index: 840860 (C4 (Web Text))

Max Activating Token Index: 8

Text #19

Max Range: 0.4117. Min Range: -0.4117

Max Act: 0.3310. Min Act: -0.0001

Data Index: 1048283 (C4 (Web Text))

Max Activating Token Index: 359