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

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

Neuron 4331 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2816. Min Act: -0.0002

Data Index: 983503 (C4 (Web Text))

Max Activating Token Index: 633

Text #1

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2697. Min Act: -0.0002

Data Index: 165658 (C4 (Web Text))

Max Activating Token Index: 761

Text #2

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2457. Min Act: -0.0002

Data Index: 348962 (C4 (Web Text))

Max Activating Token Index: 672

Text #3

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2546. Min Act: -0.0002

Data Index: 231986 (C4 (Web Text))

Max Activating Token Index: 514

Text #4

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2750. Min Act: -0.0002

Data Index: 291925 (C4 (Web Text))

Max Activating Token Index: 336

Text #5

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2850. Min Act: -0.0002

Data Index: 890387 (C4 (Web Text))

Max Activating Token Index: 364

Text #6

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2495. Min Act: -0.0002

Data Index: 110221 (C4 (Web Text))

Max Activating Token Index: 696

Text #7

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2708. Min Act: -0.0002

Data Index: 895958 (C4 (Web Text))

Max Activating Token Index: 341

Text #8

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2398. Min Act: -0.0002

Data Index: 1100827 (C4 (Web Text))

Max Activating Token Index: 749

Text #9

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2647. Min Act: -0.0002

Data Index: 757636 (C4 (Web Text))

Max Activating Token Index: 638

Text #10

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2653. Min Act: -0.0002

Data Index: 414256 (C4 (Web Text))

Max Activating Token Index: 1012

Text #11

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2878. Min Act: -0.0002

Data Index: 387875 (C4 (Web Text))

Max Activating Token Index: 687

Text #12

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2610. Min Act: -0.0002

Data Index: 43924 (C4 (Web Text))

Max Activating Token Index: 324

Text #13

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2672. Min Act: -0.0002

Data Index: 424548 (C4 (Web Text))

Max Activating Token Index: 535

Text #14

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2265. Min Act: -0.0002

Data Index: 1259583 (C4 (Web Text))

Max Activating Token Index: 124

Text #15

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2665. Min Act: -0.0002

Data Index: 606200 (C4 (Web Text))

Max Activating Token Index: 565

Text #16

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2510. Min Act: -0.0002

Data Index: 1196857 (C4 (Web Text))

Max Activating Token Index: 917

Text #17

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2585. Min Act: -0.0002

Data Index: 98191 (C4 (Web Text))

Max Activating Token Index: 928

Text #18

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2314. Min Act: -0.0002

Data Index: 991685 (C4 (Web Text))

Max Activating Token Index: 826

Text #19

Max Range: 0.2878. Min Range: -0.2878

Max Act: 0.2464. Min Act: -0.0002

Data Index: 179944 (C4 (Web Text))

Max Activating Token Index: 684