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

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

Neuron 1073 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.9470. Min Act: -0.0000

Data Index: 436994 (C4 (Web Text))

Max Activating Token Index: 120

Text #1

Max Range: 2.0092. Min Range: -2.0092

Max Act: 2.0092. Min Act: -0.0000

Data Index: 501596 (C4 (Web Text))

Max Activating Token Index: 718

Text #2

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.9016. Min Act: -0.0000

Data Index: 1539292 (Python Code)

Max Activating Token Index: 286

Text #3

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.8225. Min Act: -0.0000

Data Index: 1546667 (Python Code)

Max Activating Token Index: 883

Text #4

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.8486. Min Act: -0.0000

Data Index: 404790 (C4 (Web Text))

Max Activating Token Index: 751

Text #5

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.8200. Min Act: -0.0000

Data Index: 1408040 (Python Code)

Max Activating Token Index: 52

Text #6

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.8108. Min Act: -0.0000

Data Index: 1581621 (Python Code)

Max Activating Token Index: 26

Text #7

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.8592. Min Act: -0.0000

Data Index: 586329 (C4 (Web Text))

Max Activating Token Index: 794

Text #8

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.7978. Min Act: -0.0000

Data Index: 1196087 (C4 (Web Text))

Max Activating Token Index: 50

Text #9

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.8497. Min Act: -0.0000

Data Index: 770754 (C4 (Web Text))

Max Activating Token Index: 941

Text #10

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.7432. Min Act: -0.0000

Data Index: 1292847 (C4 (Web Text))

Max Activating Token Index: 561

Text #11

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.7624. Min Act: -0.0000

Data Index: 835392 (C4 (Web Text))

Max Activating Token Index: 625

Text #12

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.6313. Min Act: -0.0000

Data Index: 266103 (C4 (Web Text))

Max Activating Token Index: 135

Text #13

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.7317. Min Act: -0.0000

Data Index: 718101 (C4 (Web Text))

Max Activating Token Index: 976

Text #14

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.6881. Min Act: -0.0000

Data Index: 898189 (C4 (Web Text))

Max Activating Token Index: 516

Text #15

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.6836. Min Act: -0.0000

Data Index: 370024 (C4 (Web Text))

Max Activating Token Index: 775

Text #16

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.6755. Min Act: -0.0000

Data Index: 1371485 (Python Code)

Max Activating Token Index: 51

Text #17

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.6748. Min Act: -0.0000

Data Index: 607347 (C4 (Web Text))

Max Activating Token Index: 942

Text #18

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.7225. Min Act: -0.0000

Data Index: 1325326 (C4 (Web Text))

Max Activating Token Index: 500

Text #19

Max Range: 2.0092. Min Range: -2.0092

Max Act: 1.6749. Min Act: -0.0000

Data Index: 556441 (C4 (Web Text))

Max Activating Token Index: 591