Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 467 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4483. Min Act: -0.0003

Data Index: 1546940 (Python Code)

Max Activating Token Index: 1000

Text #1

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4425. Min Act: -0.0003

Data Index: 1582734 (Python Code)

Max Activating Token Index: 506

Text #2

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4528. Min Act: -0.0003

Data Index: 396253 (C4 (Web Text))

Max Activating Token Index: 420

Text #3

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4446. Min Act: -0.0003

Data Index: 1529982 (Python Code)

Max Activating Token Index: 447

Text #4

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4537. Min Act: -0.0003

Data Index: 1517412 (Python Code)

Max Activating Token Index: 291

Text #5

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4193. Min Act: -0.0003

Data Index: 1071120 (C4 (Web Text))

Max Activating Token Index: 341

Text #6

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4179. Min Act: -0.0003

Data Index: 345073 (C4 (Web Text))

Max Activating Token Index: 838

Text #7

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3885. Min Act: -0.0002

Data Index: 788627 (C4 (Web Text))

Max Activating Token Index: 640

Text #8

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4041. Min Act: -0.0003

Data Index: 1506666 (Python Code)

Max Activating Token Index: 963

Text #9

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.4048. Min Act: -0.0003

Data Index: 185541 (C4 (Web Text))

Max Activating Token Index: 940

Text #10

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3731. Min Act: -0.0003

Data Index: 525416 (C4 (Web Text))

Max Activating Token Index: 666

Text #11

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3928. Min Act: -0.0003

Data Index: 187533 (C4 (Web Text))

Max Activating Token Index: 149

Text #12

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3836. Min Act: -0.0003

Data Index: 662809 (C4 (Web Text))

Max Activating Token Index: 436

Text #13

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3915. Min Act: -0.0003

Data Index: 1529784 (Python Code)

Max Activating Token Index: 316

Text #14

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3788. Min Act: -0.0004

Data Index: 551877 (C4 (Web Text))

Max Activating Token Index: 831

Text #15

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3871. Min Act: -0.0003

Data Index: 745858 (C4 (Web Text))

Max Activating Token Index: 862

Text #16

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3852. Min Act: -0.0003

Data Index: 299635 (C4 (Web Text))

Max Activating Token Index: 103

Text #17

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3798. Min Act: -0.0003

Data Index: 1063216 (C4 (Web Text))

Max Activating Token Index: 145

Text #18

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3834. Min Act: -0.0003

Data Index: 724798 (C4 (Web Text))

Max Activating Token Index: 584

Text #19

Max Range: 0.4537. Min Range: -0.4537

Max Act: 0.3767. Min Act: -0.0003

Data Index: 1445856 (Python Code)

Max Activating Token Index: 681