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

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

Neuron 3995 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3625. Min Act: -0.0000

Data Index: 153593 (C4 (Web Text))

Max Activating Token Index: 72

Text #1

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3765. Min Act: -0.0000

Data Index: 941721 (C4 (Web Text))

Max Activating Token Index: 474

Text #2

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3659. Min Act: -0.0000

Data Index: 262121 (C4 (Web Text))

Max Activating Token Index: 913

Text #3

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3630. Min Act: -0.0000

Data Index: 992115 (C4 (Web Text))

Max Activating Token Index: 647

Text #4

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3708. Min Act: -0.0000

Data Index: 1563737 (Python Code)

Max Activating Token Index: 248

Text #5

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3494. Min Act: -0.0000

Data Index: 1183008 (C4 (Web Text))

Max Activating Token Index: 510

Text #6

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3505. Min Act: -0.0000

Data Index: 865711 (C4 (Web Text))

Max Activating Token Index: 745

Text #7

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3574. Min Act: -0.0000

Data Index: 119679 (C4 (Web Text))

Max Activating Token Index: 247

Text #8

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3247. Min Act: -0.0000

Data Index: 1320707 (C4 (Web Text))

Max Activating Token Index: 721

Text #9

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3594. Min Act: -0.0000

Data Index: 1388700 (Python Code)

Max Activating Token Index: 819

Text #10

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3298. Min Act: -0.0000

Data Index: 864629 (C4 (Web Text))

Max Activating Token Index: 219

Text #11

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3225. Min Act: -0.0000

Data Index: 569061 (C4 (Web Text))

Max Activating Token Index: 544

Text #12

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3273. Min Act: -0.0000

Data Index: 540033 (C4 (Web Text))

Max Activating Token Index: 997

Text #13

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3256. Min Act: -0.0000

Data Index: 562047 (C4 (Web Text))

Max Activating Token Index: 822

Text #14

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3242. Min Act: -0.0000

Data Index: 106865 (C4 (Web Text))

Max Activating Token Index: 390

Text #15

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3049. Min Act: -0.0000

Data Index: 1475884 (Python Code)

Max Activating Token Index: 664

Text #16

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3035. Min Act: -0.0000

Data Index: 1196005 (C4 (Web Text))

Max Activating Token Index: 783

Text #17

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3028. Min Act: -0.0000

Data Index: 778642 (C4 (Web Text))

Max Activating Token Index: 148

Text #18

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.2947. Min Act: -0.0000

Data Index: 336363 (C4 (Web Text))

Max Activating Token Index: 720

Text #19

Max Range: 0.3765. Min Range: -0.3765

Max Act: 0.3307. Min Act: -0.0000

Data Index: 1181408 (C4 (Web Text))

Max Activating Token Index: 951