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

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

Neuron 1696 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.7947. Min Act: -0.0000

Data Index: 602526 (C4 (Web Text))

Max Activating Token Index: 319

Text #1

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6724. Min Act: -0.0001

Data Index: 1209271 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6393. Min Act: -0.0001

Data Index: 1372356 (Python Code)

Max Activating Token Index: 849

Text #3

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6742. Min Act: -0.0000

Data Index: 1130381 (C4 (Web Text))

Max Activating Token Index: 739

Text #4

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6298. Min Act: -0.0000

Data Index: 295352 (C4 (Web Text))

Max Activating Token Index: 211

Text #5

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6075. Min Act: 0.0000

Data Index: 1209051 (C4 (Web Text))

Max Activating Token Index: 238

Text #6

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6302. Min Act: -0.0001

Data Index: 1518904 (Python Code)

Max Activating Token Index: 107

Text #7

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5530. Min Act: -0.0000

Data Index: 1182609 (C4 (Web Text))

Max Activating Token Index: 780

Text #8

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6031. Min Act: -0.0000

Data Index: 18076 (C4 (Web Text))

Max Activating Token Index: 662

Text #9

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5625. Min Act: -0.0000

Data Index: 669248 (C4 (Web Text))

Max Activating Token Index: 767

Text #10

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.6023. Min Act: -0.0000

Data Index: 1250087 (C4 (Web Text))

Max Activating Token Index: 467

Text #11

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5724. Min Act: -0.0000

Data Index: 253963 (C4 (Web Text))

Max Activating Token Index: 680

Text #12

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5555. Min Act: -0.0000

Data Index: 245762 (C4 (Web Text))

Max Activating Token Index: 907

Text #13

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5432. Min Act: -0.0001

Data Index: 1557562 (Python Code)

Max Activating Token Index: 680

Text #14

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5462. Min Act: -0.0000

Data Index: 1151687 (C4 (Web Text))

Max Activating Token Index: 166

Text #15

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5613. Min Act: -0.0000

Data Index: 28145 (C4 (Web Text))

Max Activating Token Index: 307

Text #16

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5183. Min Act: -0.0000

Data Index: 925459 (C4 (Web Text))

Max Activating Token Index: 542

Text #17

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5346. Min Act: -0.0000

Data Index: 1210797 (C4 (Web Text))

Max Activating Token Index: 107

Text #18

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5202. Min Act: -0.0001

Data Index: 1446635 (Python Code)

Max Activating Token Index: 556

Text #19

Max Range: 0.7947. Min Range: -0.7947

Max Act: 0.5323. Min Act: -0.0000

Data Index: 43381 (C4 (Web Text))

Max Activating Token Index: 47