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

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

Neuron 3563 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: 1.0227. Min Range: -1.0227

Max Act: 0.9568. Min Act: -0.0000

Data Index: 110387 (C4 (Web Text))

Max Activating Token Index: 741

Text #1

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8441. Min Act: -0.0000

Data Index: 373009 (C4 (Web Text))

Max Activating Token Index: 515

Text #2

Max Range: 1.0227. Min Range: -1.0227

Max Act: 1.0227. Min Act: -0.0000

Data Index: 1059510 (C4 (Web Text))

Max Activating Token Index: 987

Text #3

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.9640. Min Act: -0.0000

Data Index: 982001 (C4 (Web Text))

Max Activating Token Index: 834

Text #4

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8893. Min Act: -0.0000

Data Index: 227626 (C4 (Web Text))

Max Activating Token Index: 669

Text #5

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.9206. Min Act: -0.0000

Data Index: 1016414 (C4 (Web Text))

Max Activating Token Index: 372

Text #6

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8533. Min Act: -0.0000

Data Index: 580244 (C4 (Web Text))

Max Activating Token Index: 910

Text #7

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8597. Min Act: -0.0000

Data Index: 1040982 (C4 (Web Text))

Max Activating Token Index: 820

Text #8

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8457. Min Act: -0.0000

Data Index: 978749 (C4 (Web Text))

Max Activating Token Index: 66

Text #9

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8580. Min Act: -0.0000

Data Index: 1569739 (Python Code)

Max Activating Token Index: 510

Text #10

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8616. Min Act: -0.0000

Data Index: 1181323 (C4 (Web Text))

Max Activating Token Index: 845

Text #11

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.9176. Min Act: -0.0000

Data Index: 1339474 (C4 (Web Text))

Max Activating Token Index: 619

Text #12

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.7756. Min Act: -0.0000

Data Index: 626647 (C4 (Web Text))

Max Activating Token Index: 969

Text #13

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.7931. Min Act: -0.0000

Data Index: 433023 (C4 (Web Text))

Max Activating Token Index: 363

Text #14

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8566. Min Act: -0.0000

Data Index: 340427 (C4 (Web Text))

Max Activating Token Index: 664

Text #15

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8430. Min Act: -0.0000

Data Index: 56063 (C4 (Web Text))

Max Activating Token Index: 877

Text #16

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8156. Min Act: -0.0000

Data Index: 148077 (C4 (Web Text))

Max Activating Token Index: 261

Text #17

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8606. Min Act: -0.0000

Data Index: 719816 (C4 (Web Text))

Max Activating Token Index: 955

Text #18

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8258. Min Act: -0.0000

Data Index: 328207 (C4 (Web Text))

Max Activating Token Index: 788

Text #19

Max Range: 1.0227. Min Range: -1.0227

Max Act: 0.8307. Min Act: -0.0000

Data Index: 1038966 (C4 (Web Text))

Max Activating Token Index: 392