Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2035 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0320. Min Act: -0.0006

Data Index: 1027226 (C4 (Web Text))

Max Activating Token Index: 863

Text #1

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0308. Min Act: -0.0006

Data Index: 572616 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0292. Min Act: -0.0006

Data Index: 1078284 (C4 (Web Text))

Max Activating Token Index: 748

Text #3

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0292. Min Act: -0.0006

Data Index: 71021 (C4 (Web Text))

Max Activating Token Index: 512

Text #4

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0276. Min Act: -0.0006

Data Index: 1252100 (C4 (Web Text))

Max Activating Token Index: 139

Text #5

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0273. Min Act: -0.0006

Data Index: 501808 (C4 (Web Text))

Max Activating Token Index: 684

Text #6

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0321. Min Act: -0.0006

Data Index: 1238629 (C4 (Web Text))

Max Activating Token Index: 433

Text #7

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0288. Min Act: -0.0006

Data Index: 925944 (C4 (Web Text))

Max Activating Token Index: 421

Text #8

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0281. Min Act: -0.0006

Data Index: 1196135 (C4 (Web Text))

Max Activating Token Index: 140

Text #9

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0271. Min Act: -0.0006

Data Index: 423632 (C4 (Web Text))

Max Activating Token Index: 484

Text #10

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0276. Min Act: -0.0006

Data Index: 707404 (C4 (Web Text))

Max Activating Token Index: 621

Text #11

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0262. Min Act: -0.0006

Data Index: 774840 (C4 (Web Text))

Max Activating Token Index: 990

Text #12

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0267. Min Act: -0.0006

Data Index: 1204618 (C4 (Web Text))

Max Activating Token Index: 1009

Text #13

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0273. Min Act: -0.0006

Data Index: 190399 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0267. Min Act: -0.0006

Data Index: 1364910 (Python Code)

Max Activating Token Index: 42

Text #15

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0277. Min Act: -0.0006

Data Index: 227102 (C4 (Web Text))

Max Activating Token Index: 572

Text #16

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0265. Min Act: -0.0005

Data Index: 108707 (C4 (Web Text))

Max Activating Token Index: 384

Text #17

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0276. Min Act: -0.0006

Data Index: 1218617 (C4 (Web Text))

Max Activating Token Index: 289

Text #18

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0268. Min Act: -0.0006

Data Index: 761258 (C4 (Web Text))

Max Activating Token Index: 322

Text #19

Max Range: 0.0321. Min Range: -0.0321

Max Act: 0.0269. Min Act: -0.0006

Data Index: 1216619 (C4 (Web Text))

Max Activating Token Index: 195