Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 575 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.1090. Min Range: -1.1090

Max Act: 1.1090. Min Act: -0.0001

Data Index: 1322030 (C4 (Web Text))

Max Activating Token Index: 831

Text #1

Max Range: 1.1090. Min Range: -1.1090

Max Act: 1.0116. Min Act: -0.0001

Data Index: 1458934 (Python Code)

Max Activating Token Index: 697

Text #2

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.9830. Min Act: -0.0001

Data Index: 1609342 (Python Code)

Max Activating Token Index: 275

Text #3

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.9552. Min Act: -0.0001

Data Index: 1632418 (Python Code)

Max Activating Token Index: 893

Text #4

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.9236. Min Act: -0.0001

Data Index: 912689 (C4 (Web Text))

Max Activating Token Index: 817

Text #5

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8797. Min Act: -0.0001

Data Index: 476663 (C4 (Web Text))

Max Activating Token Index: 502

Text #6

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.9272. Min Act: -0.0001

Data Index: 99090 (C4 (Web Text))

Max Activating Token Index: 415

Text #7

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8454. Min Act: -0.0001

Data Index: 1060814 (C4 (Web Text))

Max Activating Token Index: 523

Text #8

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8480. Min Act: -0.0001

Data Index: 1654066 (Python Code)

Max Activating Token Index: 595

Text #9

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8754. Min Act: -0.0001

Data Index: 1140673 (C4 (Web Text))

Max Activating Token Index: 903

Text #10

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8429. Min Act: -0.0001

Data Index: 420220 (C4 (Web Text))

Max Activating Token Index: 861

Text #11

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8965. Min Act: -0.0001

Data Index: 56914 (C4 (Web Text))

Max Activating Token Index: 819

Text #12

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8633. Min Act: -0.0001

Data Index: 535662 (C4 (Web Text))

Max Activating Token Index: 788

Text #13

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8344. Min Act: -0.0001

Data Index: 763525 (C4 (Web Text))

Max Activating Token Index: 460

Text #14

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8844. Min Act: -0.0001

Data Index: 1069426 (C4 (Web Text))

Max Activating Token Index: 3

Text #15

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8492. Min Act: -0.0001

Data Index: 916567 (C4 (Web Text))

Max Activating Token Index: 172

Text #16

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8501. Min Act: -0.0001

Data Index: 1640407 (Python Code)

Max Activating Token Index: 703

Text #17

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8371. Min Act: -0.0001

Data Index: 953901 (C4 (Web Text))

Max Activating Token Index: 535

Text #18

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8444. Min Act: -0.0001

Data Index: 1156376 (C4 (Web Text))

Max Activating Token Index: 663

Text #19

Max Range: 1.1090. Min Range: -1.1090

Max Act: 0.8340. Min Act: -0.0001

Data Index: 399364 (C4 (Web Text))

Max Activating Token Index: 889