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

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

Neuron 3539 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: 0.1347. Min Range: -0.1347

Max Act: 0.1347. Min Act: -0.0001

Data Index: 209186 (C4 (Web Text))

Max Activating Token Index: 344

Text #1

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.1200. Min Act: -0.0001

Data Index: 862312 (C4 (Web Text))

Max Activating Token Index: 310

Text #2

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.1155. Min Act: -0.0001

Data Index: 285289 (C4 (Web Text))

Max Activating Token Index: 145

Text #3

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.1072. Min Act: -0.0001

Data Index: 294773 (C4 (Web Text))

Max Activating Token Index: 801

Text #4

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0844. Min Act: -0.0001

Data Index: 1021891 (C4 (Web Text))

Max Activating Token Index: 336

Text #5

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0765. Min Act: -0.0001

Data Index: 336005 (C4 (Web Text))

Max Activating Token Index: 148

Text #6

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0803. Min Act: -0.0001

Data Index: 296618 (C4 (Web Text))

Max Activating Token Index: 692

Text #7

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0878. Min Act: -0.0001

Data Index: 611461 (C4 (Web Text))

Max Activating Token Index: 12

Text #8

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0813. Min Act: -0.0001

Data Index: 117899 (C4 (Web Text))

Max Activating Token Index: 1015

Text #9

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0929. Min Act: -0.0001

Data Index: 515419 (C4 (Web Text))

Max Activating Token Index: 175

Text #10

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0874. Min Act: -0.0001

Data Index: 404446 (C4 (Web Text))

Max Activating Token Index: 126

Text #11

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0880. Min Act: -0.0001

Data Index: 1435907 (Python Code)

Max Activating Token Index: 299

Text #12

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0818. Min Act: -0.0001

Data Index: 1144741 (C4 (Web Text))

Max Activating Token Index: 780

Text #13

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0809. Min Act: -0.0001

Data Index: 208910 (C4 (Web Text))

Max Activating Token Index: 429

Text #14

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0800. Min Act: -0.0001

Data Index: 619228 (C4 (Web Text))

Max Activating Token Index: 75

Text #15

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0756. Min Act: -0.0001

Data Index: 1603610 (Python Code)

Max Activating Token Index: 5

Text #16

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0762. Min Act: -0.0001

Data Index: 93179 (C4 (Web Text))

Max Activating Token Index: 146

Text #17

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0755. Min Act: -0.0001

Data Index: 671847 (C4 (Web Text))

Max Activating Token Index: 364

Text #18

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0701. Min Act: -0.0001

Data Index: 1162279 (C4 (Web Text))

Max Activating Token Index: 148

Text #19

Max Range: 0.1347. Min Range: -0.1347

Max Act: 0.0736. Min Act: -0.0001

Data Index: 486375 (C4 (Web Text))

Max Activating Token Index: 200