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

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

Neuron 2812 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0162. Min Act: -0.0001

Data Index: 406343 (C4 (Web Text))

Max Activating Token Index: 773

Text #1

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0736. Min Act: -0.0001

Data Index: 130593 (C4 (Web Text))

Max Activating Token Index: 20

Text #2

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0300. Min Act: -0.0001

Data Index: 964300 (C4 (Web Text))

Max Activating Token Index: 791

Text #3

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.1616. Min Act: -0.0001

Data Index: 552223 (C4 (Web Text))

Max Activating Token Index: 537

Text #4

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.1962. Min Act: -0.0001

Data Index: 679172 (C4 (Web Text))

Max Activating Token Index: 243

Text #5

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0944. Min Act: -0.0001

Data Index: 1227165 (C4 (Web Text))

Max Activating Token Index: 999

Text #6

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0473. Min Act: -0.0001

Data Index: 1590457 (Python Code)

Max Activating Token Index: 21

Text #7

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.1069. Min Act: -0.0001

Data Index: 1129025 (C4 (Web Text))

Max Activating Token Index: 649

Text #8

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9783. Min Act: -0.0001

Data Index: 1655455 (Python Code)

Max Activating Token Index: 922

Text #9

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0021. Min Act: -0.0001

Data Index: 55144 (C4 (Web Text))

Max Activating Token Index: 619

Text #10

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0520. Min Act: -0.0001

Data Index: 738435 (C4 (Web Text))

Max Activating Token Index: 142

Text #11

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0458. Min Act: -0.0001

Data Index: 164005 (C4 (Web Text))

Max Activating Token Index: 781

Text #12

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9799. Min Act: -0.0001

Data Index: 1127025 (C4 (Web Text))

Max Activating Token Index: 337

Text #13

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9707. Min Act: -0.0001

Data Index: 46484 (C4 (Web Text))

Max Activating Token Index: 831

Text #14

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9624. Min Act: -0.0001

Data Index: 129907 (C4 (Web Text))

Max Activating Token Index: 147

Text #15

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9912. Min Act: -0.0001

Data Index: 1366877 (Python Code)

Max Activating Token Index: 137

Text #16

Max Range: 1.1962. Min Range: -1.1962

Max Act: 1.0039. Min Act: -0.0001

Data Index: 549044 (C4 (Web Text))

Max Activating Token Index: 563

Text #17

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9725. Min Act: -0.0001

Data Index: 1621842 (Python Code)

Max Activating Token Index: 27

Text #18

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.8711. Min Act: -0.0001

Data Index: 743489 (C4 (Web Text))

Max Activating Token Index: 40

Text #19

Max Range: 1.1962. Min Range: -1.1962

Max Act: 0.9963. Min Act: -0.0001

Data Index: 723190 (C4 (Web Text))

Max Activating Token Index: 110