Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4446 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.7917. Min Range: -1.7917

Max Act: 1.7917. Min Act: -0.0000

Data Index: 918138 (C4 (Web Text))

Max Activating Token Index: 12

Text #1

Max Range: 1.7917. Min Range: -1.7917

Max Act: 1.1154. Min Act: -0.0000

Data Index: 57790 (C4 (Web Text))

Max Activating Token Index: 29

Text #2

Max Range: 1.7917. Min Range: -1.7917

Max Act: 1.0461. Min Act: -0.0000

Data Index: 61788 (C4 (Web Text))

Max Activating Token Index: 1017

Text #3

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9629. Min Act: -0.0000

Data Index: 1459613 (Python Code)

Max Activating Token Index: 251

Text #4

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9508. Min Act: -0.0000

Data Index: 789298 (C4 (Web Text))

Max Activating Token Index: 153

Text #5

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9639. Min Act: -0.0000

Data Index: 1181340 (C4 (Web Text))

Max Activating Token Index: 501

Text #6

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8535. Min Act: -0.0000

Data Index: 1413339 (Python Code)

Max Activating Token Index: 909

Text #7

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8935. Min Act: -0.0000

Data Index: 1443822 (Python Code)

Max Activating Token Index: 222

Text #8

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9509. Min Act: -0.0000

Data Index: 809743 (C4 (Web Text))

Max Activating Token Index: 18

Text #9

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9016. Min Act: -0.0000

Data Index: 1528874 (Python Code)

Max Activating Token Index: 866

Text #10

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8993. Min Act: -0.0000

Data Index: 1481008 (Python Code)

Max Activating Token Index: 157

Text #11

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9354. Min Act: -0.0000

Data Index: 1391574 (Python Code)

Max Activating Token Index: 200

Text #12

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9041. Min Act: -0.0000

Data Index: 1010218 (C4 (Web Text))

Max Activating Token Index: 976

Text #13

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9226. Min Act: -0.0000

Data Index: 914259 (C4 (Web Text))

Max Activating Token Index: 823

Text #14

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8623. Min Act: -0.0000

Data Index: 980622 (C4 (Web Text))

Max Activating Token Index: 689

Text #15

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8619. Min Act: -0.0000

Data Index: 586439 (C4 (Web Text))

Max Activating Token Index: 195

Text #16

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.9116. Min Act: -0.0000

Data Index: 1561137 (Python Code)

Max Activating Token Index: 625

Text #17

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8817. Min Act: -0.0000

Data Index: 937390 (C4 (Web Text))

Max Activating Token Index: 880

Text #18

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8150. Min Act: -0.0000

Data Index: 1562388 (Python Code)

Max Activating Token Index: 796

Text #19

Max Range: 1.7917. Min Range: -1.7917

Max Act: 0.8642. Min Act: -0.0000

Data Index: 96937 (C4 (Web Text))

Max Activating Token Index: 332