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

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

Neuron 478 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: 0.0386. Min Range: -0.0386

Max Act: 0.0386. Min Act: -0.0001

Data Index: 295909 (C4 (Web Text))

Max Activating Token Index: 889

Text #1

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0327. Min Act: -0.0000

Data Index: 1654700 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0348. Min Act: -0.0001

Data Index: 1441739 (Python Code)

Max Activating Token Index: 386

Text #3

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0323. Min Act: -0.0001

Data Index: 1039373 (C4 (Web Text))

Max Activating Token Index: 199

Text #4

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0305. Min Act: -0.0001

Data Index: 201452 (C4 (Web Text))

Max Activating Token Index: 342

Text #5

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0312. Min Act: -0.0001

Data Index: 1130609 (C4 (Web Text))

Max Activating Token Index: 919

Text #6

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0259. Min Act: -0.0001

Data Index: 532559 (C4 (Web Text))

Max Activating Token Index: 321

Text #7

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0259. Min Act: -0.0000

Data Index: 874974 (C4 (Web Text))

Max Activating Token Index: 313

Text #8

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0258. Min Act: -0.0001

Data Index: 1482598 (Python Code)

Max Activating Token Index: 30

Text #9

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0261. Min Act: -0.0000

Data Index: 62727 (C4 (Web Text))

Max Activating Token Index: 1001

Text #10

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0251. Min Act: -0.0001

Data Index: 313877 (C4 (Web Text))

Max Activating Token Index: 417

Text #11

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0246. Min Act: -0.0000

Data Index: 39262 (C4 (Web Text))

Max Activating Token Index: 675

Text #12

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0251. Min Act: -0.0001

Data Index: 917628 (C4 (Web Text))

Max Activating Token Index: 29

Text #13

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0251. Min Act: -0.0000

Data Index: 23877 (C4 (Web Text))

Max Activating Token Index: 958

Text #14

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0237. Min Act: -0.0001

Data Index: 1233587 (C4 (Web Text))

Max Activating Token Index: 864

Text #15

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0241. Min Act: -0.0001

Data Index: 1519493 (Python Code)

Max Activating Token Index: 112

Text #16

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0227. Min Act: -0.0001

Data Index: 229504 (C4 (Web Text))

Max Activating Token Index: 558

Text #17

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0246. Min Act: -0.0001

Data Index: 977926 (C4 (Web Text))

Max Activating Token Index: 893

Text #18

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0234. Min Act: -0.0001

Data Index: 1017112 (C4 (Web Text))

Max Activating Token Index: 748

Text #19

Max Range: 0.0386. Min Range: -0.0386

Max Act: 0.0249. Min Act: -0.0001

Data Index: 767034 (C4 (Web Text))

Max Activating Token Index: 94