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

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

Neuron 3494 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 7.4418. Min Range: -7.4418

Max Act: 7.4418. Min Act: -0.0001

Data Index: 1365094 (Python Code)

Max Activating Token Index: 271

Text #1

Max Range: 7.4418. Min Range: -7.4418

Max Act: 7.0206. Min Act: -0.0001

Data Index: 1372957 (Python Code)

Max Activating Token Index: 954

Text #2

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.7168. Min Act: -0.0001

Data Index: 1439077 (Python Code)

Max Activating Token Index: 1012

Text #3

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.2309. Min Act: -0.0001

Data Index: 1162927 (C4 (Web Text))

Max Activating Token Index: 793

Text #4

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.1970. Min Act: -0.0001

Data Index: 159344 (C4 (Web Text))

Max Activating Token Index: 634

Text #5

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.1475. Min Act: -0.0001

Data Index: 585220 (C4 (Web Text))

Max Activating Token Index: 308

Text #6

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.0629. Min Act: -0.0001

Data Index: 1644708 (Python Code)

Max Activating Token Index: 397

Text #7

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.9452. Min Act: -0.0001

Data Index: 1469573 (Python Code)

Max Activating Token Index: 434

Text #8

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.1038. Min Act: -0.0001

Data Index: 1596966 (Python Code)

Max Activating Token Index: 22

Text #9

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.9053. Min Act: -0.0001

Data Index: 1498990 (Python Code)

Max Activating Token Index: 833

Text #10

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.2250. Min Act: -0.0001

Data Index: 888188 (C4 (Web Text))

Max Activating Token Index: 450

Text #11

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.6660. Min Act: -0.0001

Data Index: 601794 (C4 (Web Text))

Max Activating Token Index: 595

Text #12

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.1519. Min Act: -0.0001

Data Index: 1589399 (Python Code)

Max Activating Token Index: 582

Text #13

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.3129. Min Act: -0.0001

Data Index: 1545153 (Python Code)

Max Activating Token Index: 704

Text #14

Max Range: 7.4418. Min Range: -7.4418

Max Act: 6.1436. Min Act: -0.0001

Data Index: 804080 (C4 (Web Text))

Max Activating Token Index: 586

Text #15

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.9479. Min Act: -0.0001

Data Index: 1192717 (C4 (Web Text))

Max Activating Token Index: 196

Text #16

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.9400. Min Act: -0.0001

Data Index: 1582183 (Python Code)

Max Activating Token Index: 291

Text #17

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.5400. Min Act: -0.0001

Data Index: 1585848 (Python Code)

Max Activating Token Index: 144

Text #18

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.7023. Min Act: -0.0001

Data Index: 1434444 (Python Code)

Max Activating Token Index: 779

Text #19

Max Range: 7.4418. Min Range: -7.4418

Max Act: 5.6706. Min Act: -0.0001

Data Index: 406631 (C4 (Web Text))

Max Activating Token Index: 181