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

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

Neuron 1830 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.9358. Min Act: -0.0000

Data Index: 1151018 (C4 (Web Text))

Max Activating Token Index: 990

Text #1

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.9407. Min Act: -0.0000

Data Index: 78067 (C4 (Web Text))

Max Activating Token Index: 819

Text #2

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.7723. Min Act: -0.0000

Data Index: 204689 (C4 (Web Text))

Max Activating Token Index: 401

Text #3

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.7362. Min Act: -0.0000

Data Index: 1184738 (C4 (Web Text))

Max Activating Token Index: 867

Text #4

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.5353. Min Act: -0.0000

Data Index: 657823 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.2213. Min Act: -0.0000

Data Index: 1006723 (C4 (Web Text))

Max Activating Token Index: 696

Text #6

Max Range: 4.9407. Min Range: -4.9407

Max Act: 4.1710. Min Act: -0.0000

Data Index: 599905 (C4 (Web Text))

Max Activating Token Index: 587

Text #7

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.7614. Min Act: -0.0000

Data Index: 877515 (C4 (Web Text))

Max Activating Token Index: 903

Text #8

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.9061. Min Act: -0.0000

Data Index: 331380 (C4 (Web Text))

Max Activating Token Index: 385

Text #9

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.8671. Min Act: -0.0000

Data Index: 645893 (C4 (Web Text))

Max Activating Token Index: 782

Text #10

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.8188. Min Act: -0.0000

Data Index: 351474 (C4 (Web Text))

Max Activating Token Index: 761

Text #11

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.8371. Min Act: -0.0000

Data Index: 1063286 (C4 (Web Text))

Max Activating Token Index: 592

Text #12

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.8363. Min Act: -0.0000

Data Index: 995805 (C4 (Web Text))

Max Activating Token Index: 964

Text #13

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.7111. Min Act: -0.0000

Data Index: 703051 (C4 (Web Text))

Max Activating Token Index: 835

Text #14

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.9753. Min Act: -0.0000

Data Index: 692342 (C4 (Web Text))

Max Activating Token Index: 1017

Text #15

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.7589. Min Act: -0.0000

Data Index: 989438 (C4 (Web Text))

Max Activating Token Index: 609

Text #16

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.4917. Min Act: -0.0000

Data Index: 1097020 (C4 (Web Text))

Max Activating Token Index: 677

Text #17

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.6246. Min Act: -0.0000

Data Index: 308621 (C4 (Web Text))

Max Activating Token Index: 572

Text #18

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.6269. Min Act: -0.0000

Data Index: 714079 (C4 (Web Text))

Max Activating Token Index: 933

Text #19

Max Range: 4.9407. Min Range: -4.9407

Max Act: 3.7221. Min Act: -0.0000

Data Index: 523626 (C4 (Web Text))

Max Activating Token Index: 230