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

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

Neuron 1984 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.3511. Min Act: -0.0001

Data Index: 146532 (C4 (Web Text))

Max Activating Token Index: 460

Text #1

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.3239. Min Act: -0.0001

Data Index: 940580 (C4 (Web Text))

Max Activating Token Index: 820

Text #2

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2815. Min Act: -0.0001

Data Index: 99381 (C4 (Web Text))

Max Activating Token Index: 7

Text #3

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2779. Min Act: -0.0001

Data Index: 1430240 (Python Code)

Max Activating Token Index: 442

Text #4

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2515. Min Act: -0.0001

Data Index: 923309 (C4 (Web Text))

Max Activating Token Index: 628

Text #5

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2725. Min Act: -0.0001

Data Index: 301755 (C4 (Web Text))

Max Activating Token Index: 9

Text #6

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2800. Min Act: -0.0001

Data Index: 1027377 (C4 (Web Text))

Max Activating Token Index: 658

Text #7

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.3035. Min Act: -0.0001

Data Index: 694157 (C4 (Web Text))

Max Activating Token Index: 841

Text #8

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2794. Min Act: -0.0001

Data Index: 83647 (C4 (Web Text))

Max Activating Token Index: 196

Text #9

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2820. Min Act: -0.0001

Data Index: 153476 (C4 (Web Text))

Max Activating Token Index: 985

Text #10

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2609. Min Act: -0.0001

Data Index: 747252 (C4 (Web Text))

Max Activating Token Index: 180

Text #11

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2481. Min Act: -0.0001

Data Index: 1433520 (Python Code)

Max Activating Token Index: 920

Text #12

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2723. Min Act: -0.0001

Data Index: 913500 (C4 (Web Text))

Max Activating Token Index: 931

Text #13

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2520. Min Act: -0.0001

Data Index: 1226400 (C4 (Web Text))

Max Activating Token Index: 801

Text #14

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2499. Min Act: -0.0001

Data Index: 1260092 (C4 (Web Text))

Max Activating Token Index: 19

Text #15

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2531. Min Act: -0.0001

Data Index: 1526048 (Python Code)

Max Activating Token Index: 565

Text #16

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2854. Min Act: -0.0001

Data Index: 1388377 (Python Code)

Max Activating Token Index: 226

Text #17

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2571. Min Act: -0.0001

Data Index: 1624798 (Python Code)

Max Activating Token Index: 665

Text #18

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2586. Min Act: -0.0001

Data Index: 1469003 (Python Code)

Max Activating Token Index: 905

Text #19

Max Range: 0.3511. Min Range: -0.3511

Max Act: 0.2582. Min Act: -0.0001

Data Index: 868463 (C4 (Web Text))

Max Activating Token Index: 931