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

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

Neuron 4279 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: 0.3898. Min Range: -0.3898

Max Act: 0.3898. Min Act: 0.0000

Data Index: 721671 (C4 (Web Text))

Max Activating Token Index: 733

Text #1

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.3503. Min Act: -0.0000

Data Index: 1650092 (Python Code)

Max Activating Token Index: 1010

Text #2

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.3434. Min Act: -0.0000

Data Index: 1523330 (Python Code)

Max Activating Token Index: 937

Text #3

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.3398. Min Act: -0.0000

Data Index: 1123980 (C4 (Web Text))

Max Activating Token Index: 914

Text #4

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.3086. Min Act: -0.0000

Data Index: 1409335 (Python Code)

Max Activating Token Index: 451

Text #5

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.3030. Min Act: -0.0000

Data Index: 1516898 (Python Code)

Max Activating Token Index: 487

Text #6

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.3062. Min Act: -0.0000

Data Index: 1191238 (C4 (Web Text))

Max Activating Token Index: 243

Text #7

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2854. Min Act: -0.0000

Data Index: 1177903 (C4 (Web Text))

Max Activating Token Index: 483

Text #8

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2573. Min Act: -0.0000

Data Index: 611699 (C4 (Web Text))

Max Activating Token Index: 401

Text #9

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2632. Min Act: 0.0000

Data Index: 1251566 (C4 (Web Text))

Max Activating Token Index: 596

Text #10

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2588. Min Act: -0.0000

Data Index: 1643284 (Python Code)

Max Activating Token Index: 477

Text #11

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2748. Min Act: -0.0000

Data Index: 415393 (C4 (Web Text))

Max Activating Token Index: 679

Text #12

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2666. Min Act: -0.0000

Data Index: 1094092 (C4 (Web Text))

Max Activating Token Index: 696

Text #13

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2509. Min Act: -0.0000

Data Index: 564776 (C4 (Web Text))

Max Activating Token Index: 1019

Text #14

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2447. Min Act: -0.0000

Data Index: 123606 (C4 (Web Text))

Max Activating Token Index: 201

Text #15

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2538. Min Act: 0.0000

Data Index: 1310783 (C4 (Web Text))

Max Activating Token Index: 362

Text #16

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2364. Min Act: -0.0000

Data Index: 1578325 (Python Code)

Max Activating Token Index: 107

Text #17

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2386. Min Act: -0.0000

Data Index: 234580 (C4 (Web Text))

Max Activating Token Index: 767

Text #18

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2303. Min Act: -0.0000

Data Index: 1595708 (Python Code)

Max Activating Token Index: 382

Text #19

Max Range: 0.3898. Min Range: -0.3898

Max Act: 0.2282. Min Act: -0.0000

Data Index: 144797 (C4 (Web Text))

Max Activating Token Index: 666