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

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

Neuron 4558 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.6865. Min Act: -0.0000

Data Index: 1192702 (C4 (Web Text))

Max Activating Token Index: 930

Text #1

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.5190. Min Act: -0.0000

Data Index: 1441670 (Python Code)

Max Activating Token Index: 215

Text #2

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.4910. Min Act: -0.0000

Data Index: 1109426 (C4 (Web Text))

Max Activating Token Index: 548

Text #3

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.4354. Min Act: -0.0000

Data Index: 317619 (C4 (Web Text))

Max Activating Token Index: 967

Text #4

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.3285. Min Act: -0.0000

Data Index: 1491552 (Python Code)

Max Activating Token Index: 29

Text #5

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2514. Min Act: -0.0000

Data Index: 258820 (C4 (Web Text))

Max Activating Token Index: 732

Text #6

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2284. Min Act: -0.0000

Data Index: 126708 (C4 (Web Text))

Max Activating Token Index: 189

Text #7

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2863. Min Act: -0.0000

Data Index: 335142 (C4 (Web Text))

Max Activating Token Index: 202

Text #8

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2265. Min Act: -0.0000

Data Index: 1355120 (C4 (Web Text))

Max Activating Token Index: 387

Text #9

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2507. Min Act: -0.0000

Data Index: 1629072 (Python Code)

Max Activating Token Index: 54

Text #10

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2045. Min Act: -0.0000

Data Index: 1568559 (Python Code)

Max Activating Token Index: 187

Text #11

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.2236. Min Act: -0.0000

Data Index: 1444458 (Python Code)

Max Activating Token Index: 867

Text #12

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.1577. Min Act: -0.0000

Data Index: 1262669 (C4 (Web Text))

Max Activating Token Index: 310

Text #13

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.1015. Min Act: -0.0000

Data Index: 1576199 (Python Code)

Max Activating Token Index: 881

Text #14

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.1170. Min Act: -0.0000

Data Index: 78203 (C4 (Web Text))

Max Activating Token Index: 142

Text #15

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.0230. Min Act: -0.0000

Data Index: 1423406 (Python Code)

Max Activating Token Index: 64

Text #16

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.0510. Min Act: -0.0000

Data Index: 1516404 (Python Code)

Max Activating Token Index: 700

Text #17

Max Range: 1.6865. Min Range: -1.6865

Max Act: 0.9851. Min Act: -0.0000

Data Index: 1634046 (Python Code)

Max Activating Token Index: 51

Text #18

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.0725. Min Act: -0.0000

Data Index: 1212166 (C4 (Web Text))

Max Activating Token Index: 267

Text #19

Max Range: 1.6865. Min Range: -1.6865

Max Act: 1.0467. Min Act: -0.0000

Data Index: 1513015 (Python Code)

Max Activating Token Index: 710