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

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

Neuron 4597 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.1473. Min Range: -2.1473

Max Act: 2.1473. Min Act: -0.0000

Data Index: 729828 (C4 (Web Text))

Max Activating Token Index: 717

Text #1

Max Range: 2.1473. Min Range: -2.1473

Max Act: 2.0218. Min Act: -0.0000

Data Index: 1261602 (C4 (Web Text))

Max Activating Token Index: 746

Text #2

Max Range: 2.1473. Min Range: -2.1473

Max Act: 2.0264. Min Act: -0.0000

Data Index: 1581113 (Python Code)

Max Activating Token Index: 695

Text #3

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.9655. Min Act: -0.0000

Data Index: 406645 (C4 (Web Text))

Max Activating Token Index: 785

Text #4

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.9271. Min Act: -0.0000

Data Index: 702572 (C4 (Web Text))

Max Activating Token Index: 1013

Text #5

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.7258. Min Act: -0.0000

Data Index: 1469693 (Python Code)

Max Activating Token Index: 758

Text #6

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.7331. Min Act: -0.0000

Data Index: 1600709 (Python Code)

Max Activating Token Index: 705

Text #7

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.7027. Min Act: -0.0000

Data Index: 1524244 (Python Code)

Max Activating Token Index: 398

Text #8

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.5994. Min Act: -0.0000

Data Index: 1213119 (C4 (Web Text))

Max Activating Token Index: 62

Text #9

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.5665. Min Act: -0.0000

Data Index: 1058671 (C4 (Web Text))

Max Activating Token Index: 204

Text #10

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.6727. Min Act: -0.0000

Data Index: 466022 (C4 (Web Text))

Max Activating Token Index: 559

Text #11

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.5867. Min Act: -0.0000

Data Index: 1589082 (Python Code)

Max Activating Token Index: 422

Text #12

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.5365. Min Act: -0.0000

Data Index: 1622823 (Python Code)

Max Activating Token Index: 999

Text #13

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.6035. Min Act: -0.0000

Data Index: 638563 (C4 (Web Text))

Max Activating Token Index: 299

Text #14

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.5792. Min Act: -0.0000

Data Index: 1261557 (C4 (Web Text))

Max Activating Token Index: 226

Text #15

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.4501. Min Act: -0.0000

Data Index: 1606326 (Python Code)

Max Activating Token Index: 342

Text #16

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.5576. Min Act: -0.0000

Data Index: 1145901 (C4 (Web Text))

Max Activating Token Index: 877

Text #17

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.6118. Min Act: -0.0000

Data Index: 1637220 (Python Code)

Max Activating Token Index: 300

Text #18

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.4029. Min Act: -0.0000

Data Index: 311777 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 2.1473. Min Range: -2.1473

Max Act: 1.4150. Min Act: -0.0000

Data Index: 1024097 (C4 (Web Text))

Max Activating Token Index: 519