Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: The Pile

Neuron 4769 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 7.8325. Min Range: -7.8325

Max Act: 7.8325. Min Act: -0.0000

Data Index: 1963192 (The Pile)

Max Activating Token Index: 834

Text #1

Max Range: 7.8325. Min Range: -7.8325

Max Act: 7.7330. Min Act: -0.0000

Data Index: 839900 (The Pile)

Max Activating Token Index: 597

Text #2

Max Range: 7.8325. Min Range: -7.8325

Max Act: 7.6068. Min Act: -0.0000

Data Index: 651500 (The Pile)

Max Activating Token Index: 303

Text #3

Max Range: 7.8325. Min Range: -7.8325

Max Act: 7.0449. Min Act: -0.0000

Data Index: 215832 (The Pile)

Max Activating Token Index: 199

Text #4

Max Range: 7.8325. Min Range: -7.8325

Max Act: 7.2580. Min Act: -0.0000

Data Index: 604916 (The Pile)

Max Activating Token Index: 23

Text #5

Max Range: 7.8325. Min Range: -7.8325

Max Act: 7.2779. Min Act: -0.0000

Data Index: 1714415 (The Pile)

Max Activating Token Index: 222

Text #6

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.6513. Min Act: -0.0000

Data Index: 1886861 (The Pile)

Max Activating Token Index: 700

Text #7

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.5582. Min Act: -0.0000

Data Index: 1403848 (The Pile)

Max Activating Token Index: 65

Text #8

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.3201. Min Act: -0.0000

Data Index: 1083510 (The Pile)

Max Activating Token Index: 208

Text #9

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.4563. Min Act: -0.0000

Data Index: 1162849 (The Pile)

Max Activating Token Index: 121

Text #10

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.9739. Min Act: -0.0000

Data Index: 387862 (The Pile)

Max Activating Token Index: 138

Text #11

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.9388. Min Act: -0.0000

Data Index: 1101007 (The Pile)

Max Activating Token Index: 268

Text #12

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.1270. Min Act: -0.0000

Data Index: 301274 (The Pile)

Max Activating Token Index: 97

Text #13

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.0911. Min Act: -0.0000

Data Index: 629689 (The Pile)

Max Activating Token Index: 439

Text #14

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.9279. Min Act: -0.0000

Data Index: 1167774 (The Pile)

Max Activating Token Index: 148

Text #15

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.5319. Min Act: -0.0000

Data Index: 654415 (The Pile)

Max Activating Token Index: 813

Text #16

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.9454. Min Act: -0.0000

Data Index: 176561 (The Pile)

Max Activating Token Index: 229

Text #17

Max Range: 7.8325. Min Range: -7.8325

Max Act: 6.4428. Min Act: -0.0000

Data Index: 1613394 (The Pile)

Max Activating Token Index: 365

Text #18

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.8313. Min Act: -0.0000

Data Index: 594440 (The Pile)

Max Activating Token Index: 35

Text #19

Max Range: 7.8325. Min Range: -7.8325

Max Act: 5.0646. Min Act: -0.0000

Data Index: 719866 (The Pile)

Max Activating Token Index: 787