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

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

Neuron 550 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.6481. Min Act: -0.0000

Data Index: 815037 (C4 (Web Text))

Max Activating Token Index: 323

Text #1

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.6498. Min Act: -0.0000

Data Index: 910223 (C4 (Web Text))

Max Activating Token Index: 531

Text #2

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.6273. Min Act: -0.0000

Data Index: 887388 (C4 (Web Text))

Max Activating Token Index: 165

Text #3

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.6133. Min Act: -0.0000

Data Index: 611015 (C4 (Web Text))

Max Activating Token Index: 381

Text #4

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5982. Min Act: -0.0000

Data Index: 826526 (C4 (Web Text))

Max Activating Token Index: 888

Text #5

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5684. Min Act: -0.0000

Data Index: 233720 (C4 (Web Text))

Max Activating Token Index: 139

Text #6

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5767. Min Act: -0.0000

Data Index: 555728 (C4 (Web Text))

Max Activating Token Index: 1020

Text #7

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5305. Min Act: -0.0000

Data Index: 542464 (C4 (Web Text))

Max Activating Token Index: 34

Text #8

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5401. Min Act: -0.0000

Data Index: 298961 (C4 (Web Text))

Max Activating Token Index: 976

Text #9

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5413. Min Act: -0.0000

Data Index: 418924 (C4 (Web Text))

Max Activating Token Index: 487

Text #10

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5427. Min Act: -0.0000

Data Index: 1090235 (C4 (Web Text))

Max Activating Token Index: 117

Text #11

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5145. Min Act: -0.0000

Data Index: 13571 (C4 (Web Text))

Max Activating Token Index: 433

Text #12

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5082. Min Act: -0.0000

Data Index: 576059 (C4 (Web Text))

Max Activating Token Index: 98

Text #13

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5119. Min Act: -0.0000

Data Index: 1438417 (Python Code)

Max Activating Token Index: 720

Text #14

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.4811. Min Act: -0.0000

Data Index: 1357120 (C4 (Web Text))

Max Activating Token Index: 956

Text #15

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5000. Min Act: 0.0000

Data Index: 1399123 (Python Code)

Max Activating Token Index: 415

Text #16

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.5047. Min Act: -0.0000

Data Index: 1434596 (Python Code)

Max Activating Token Index: 867

Text #17

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.4478. Min Act: -0.0000

Data Index: 1229990 (C4 (Web Text))

Max Activating Token Index: 489

Text #18

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.4847. Min Act: -0.0000

Data Index: 862895 (C4 (Web Text))

Max Activating Token Index: 150

Text #19

Max Range: 0.6498. Min Range: -0.6498

Max Act: 0.4519. Min Act: -0.0000

Data Index: 1016424 (C4 (Web Text))

Max Activating Token Index: 332