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

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

Neuron 4827 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.8820. Min Act: -0.0000

Data Index: 627880 (C4 (Web Text))

Max Activating Token Index: 319

Text #1

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.7168. Min Act: -0.0000

Data Index: 1209295 (C4 (Web Text))

Max Activating Token Index: 875

Text #2

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.6139. Min Act: -0.0000

Data Index: 553815 (C4 (Web Text))

Max Activating Token Index: 205

Text #3

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.4930. Min Act: -0.0000

Data Index: 1468544 (Python Code)

Max Activating Token Index: 414

Text #4

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.4155. Min Act: -0.0000

Data Index: 1225113 (C4 (Web Text))

Max Activating Token Index: 87

Text #5

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.4446. Min Act: -0.0000

Data Index: 1321282 (C4 (Web Text))

Max Activating Token Index: 127

Text #6

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.4753. Min Act: -0.0000

Data Index: 535352 (C4 (Web Text))

Max Activating Token Index: 1007

Text #7

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3816. Min Act: -0.0000

Data Index: 911901 (C4 (Web Text))

Max Activating Token Index: 356

Text #8

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.4630. Min Act: -0.0000

Data Index: 793504 (C4 (Web Text))

Max Activating Token Index: 96

Text #9

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3971. Min Act: -0.0000

Data Index: 1470589 (Python Code)

Max Activating Token Index: 105

Text #10

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3431. Min Act: -0.0000

Data Index: 464659 (C4 (Web Text))

Max Activating Token Index: 741

Text #11

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.2580. Min Act: -0.0000

Data Index: 522633 (C4 (Web Text))

Max Activating Token Index: 791

Text #12

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3292. Min Act: -0.0000

Data Index: 1399204 (Python Code)

Max Activating Token Index: 402

Text #13

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.2744. Min Act: -0.0000

Data Index: 1420851 (Python Code)

Max Activating Token Index: 328

Text #14

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3818. Min Act: -0.0000

Data Index: 815594 (C4 (Web Text))

Max Activating Token Index: 834

Text #15

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3311. Min Act: -0.0000

Data Index: 1608829 (Python Code)

Max Activating Token Index: 111

Text #16

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.2922. Min Act: -0.0000

Data Index: 111582 (C4 (Web Text))

Max Activating Token Index: 316

Text #17

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.3242. Min Act: -0.0000

Data Index: 1541631 (Python Code)

Max Activating Token Index: 496

Text #18

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.2247. Min Act: -0.0000

Data Index: 1477980 (Python Code)

Max Activating Token Index: 607

Text #19

Max Range: 1.8820. Min Range: -1.8820

Max Act: 1.2576. Min Act: -0.0000

Data Index: 1554739 (Python Code)

Max Activating Token Index: 56