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

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

Neuron 426 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2949. Min Act: -0.0001

Data Index: 665593 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2960. Min Act: -0.0001

Data Index: 255593 (C4 (Web Text))

Max Activating Token Index: 153

Text #2

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2592. Min Act: -0.0001

Data Index: 742617 (C4 (Web Text))

Max Activating Token Index: 667

Text #3

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2692. Min Act: -0.0001

Data Index: 937147 (C4 (Web Text))

Max Activating Token Index: 17

Text #4

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2323. Min Act: -0.0001

Data Index: 840997 (C4 (Web Text))

Max Activating Token Index: 208

Text #5

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2376. Min Act: -0.0002

Data Index: 1311458 (C4 (Web Text))

Max Activating Token Index: 70

Text #6

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2370. Min Act: -0.0001

Data Index: 89943 (C4 (Web Text))

Max Activating Token Index: 268

Text #7

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2454. Min Act: -0.0001

Data Index: 496449 (C4 (Web Text))

Max Activating Token Index: 707

Text #8

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2714. Min Act: -0.0001

Data Index: 472261 (C4 (Web Text))

Max Activating Token Index: 508

Text #9

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2417. Min Act: -0.0002

Data Index: 1321579 (C4 (Web Text))

Max Activating Token Index: 627

Text #10

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2409. Min Act: -0.0001

Data Index: 624879 (C4 (Web Text))

Max Activating Token Index: 218

Text #11

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2303. Min Act: -0.0001

Data Index: 726770 (C4 (Web Text))

Max Activating Token Index: 650

Text #12

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2248. Min Act: -0.0001

Data Index: 1314636 (C4 (Web Text))

Max Activating Token Index: 274

Text #13

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2356. Min Act: -0.0001

Data Index: 658990 (C4 (Web Text))

Max Activating Token Index: 418

Text #14

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2597. Min Act: -0.0001

Data Index: 1286500 (C4 (Web Text))

Max Activating Token Index: 94

Text #15

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2303. Min Act: -0.0001

Data Index: 583222 (C4 (Web Text))

Max Activating Token Index: 285

Text #16

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2435. Min Act: -0.0002

Data Index: 710445 (C4 (Web Text))

Max Activating Token Index: 620

Text #17

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2343. Min Act: -0.0002

Data Index: 313192 (C4 (Web Text))

Max Activating Token Index: 19

Text #18

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2270. Min Act: -0.0001

Data Index: 766283 (C4 (Web Text))

Max Activating Token Index: 675

Text #19

Max Range: 0.2960. Min Range: -0.2960

Max Act: 0.2227. Min Act: -0.0001

Data Index: 1204927 (C4 (Web Text))

Max Activating Token Index: 214