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

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

Neuron 18 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.7989. Min Act: -0.0001

Data Index: 877695 (C4 (Web Text))

Max Activating Token Index: 310

Text #1

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.8039. Min Act: -0.0001

Data Index: 333109 (C4 (Web Text))

Max Activating Token Index: 16

Text #2

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.6607. Min Act: -0.0001

Data Index: 436630 (C4 (Web Text))

Max Activating Token Index: 536

Text #3

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.6687. Min Act: -0.0001

Data Index: 866948 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5826. Min Act: -0.0001

Data Index: 874328 (C4 (Web Text))

Max Activating Token Index: 952

Text #5

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5997. Min Act: -0.0001

Data Index: 190861 (C4 (Web Text))

Max Activating Token Index: 52

Text #6

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5746. Min Act: -0.0001

Data Index: 1007152 (C4 (Web Text))

Max Activating Token Index: 389

Text #7

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5659. Min Act: -0.0001

Data Index: 292545 (C4 (Web Text))

Max Activating Token Index: 877

Text #8

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5755. Min Act: -0.0001

Data Index: 1260155 (C4 (Web Text))

Max Activating Token Index: 883

Text #9

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5557. Min Act: -0.0001

Data Index: 1083484 (C4 (Web Text))

Max Activating Token Index: 338

Text #10

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5706. Min Act: -0.0001

Data Index: 1049000 (C4 (Web Text))

Max Activating Token Index: 300

Text #11

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5428. Min Act: -0.0001

Data Index: 1177422 (C4 (Web Text))

Max Activating Token Index: 74

Text #12

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5303. Min Act: -0.0001

Data Index: 231704 (C4 (Web Text))

Max Activating Token Index: 401

Text #13

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5278. Min Act: -0.0001

Data Index: 1224727 (C4 (Web Text))

Max Activating Token Index: 749

Text #14

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5581. Min Act: -0.0001

Data Index: 425547 (C4 (Web Text))

Max Activating Token Index: 837

Text #15

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5633. Min Act: -0.0001

Data Index: 1338482 (C4 (Web Text))

Max Activating Token Index: 670

Text #16

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.4753. Min Act: -0.0001

Data Index: 1273383 (C4 (Web Text))

Max Activating Token Index: 307

Text #17

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5623. Min Act: -0.0001

Data Index: 120436 (C4 (Web Text))

Max Activating Token Index: 1009

Text #18

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5959. Min Act: -0.0001

Data Index: 569968 (C4 (Web Text))

Max Activating Token Index: 941

Text #19

Max Range: 0.8039. Min Range: -0.8039

Max Act: 0.5400. Min Act: -0.0001

Data Index: 840714 (C4 (Web Text))

Max Activating Token Index: 872