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

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

Neuron 4823 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.8592. Min Act: -0.0000

Data Index: 168458 (C4 (Web Text))

Max Activating Token Index: 277

Text #1

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.8267. Min Act: -0.0000

Data Index: 1399951 (Python Code)

Max Activating Token Index: 771

Text #2

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.7780. Min Act: -0.0000

Data Index: 403252 (C4 (Web Text))

Max Activating Token Index: 522

Text #3

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.7733. Min Act: -0.0000

Data Index: 308507 (C4 (Web Text))

Max Activating Token Index: 159

Text #4

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6915. Min Act: -0.0000

Data Index: 908582 (C4 (Web Text))

Max Activating Token Index: 665

Text #5

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.7679. Min Act: -0.0000

Data Index: 1380661 (Python Code)

Max Activating Token Index: 91

Text #6

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6588. Min Act: -0.0000

Data Index: 1161766 (C4 (Web Text))

Max Activating Token Index: 653

Text #7

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.7394. Min Act: -0.0000

Data Index: 432323 (C4 (Web Text))

Max Activating Token Index: 564

Text #8

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6949. Min Act: -0.0000

Data Index: 1456292 (Python Code)

Max Activating Token Index: 74

Text #9

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6852. Min Act: -0.0000

Data Index: 1612051 (Python Code)

Max Activating Token Index: 966

Text #10

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6655. Min Act: -0.0000

Data Index: 1605940 (Python Code)

Max Activating Token Index: 146

Text #11

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6649. Min Act: -0.0000

Data Index: 1497548 (Python Code)

Max Activating Token Index: 140

Text #12

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6980. Min Act: -0.0000

Data Index: 202304 (C4 (Web Text))

Max Activating Token Index: 765

Text #13

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6359. Min Act: -0.0000

Data Index: 1405729 (Python Code)

Max Activating Token Index: 742

Text #14

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6774. Min Act: -0.0000

Data Index: 137305 (C4 (Web Text))

Max Activating Token Index: 795

Text #15

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6521. Min Act: -0.0000

Data Index: 937729 (C4 (Web Text))

Max Activating Token Index: 413

Text #16

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6506. Min Act: -0.0000

Data Index: 1623533 (Python Code)

Max Activating Token Index: 163

Text #17

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6494. Min Act: -0.0000

Data Index: 1466684 (Python Code)

Max Activating Token Index: 129

Text #18

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6297. Min Act: -0.0000

Data Index: 1593342 (Python Code)

Max Activating Token Index: 107

Text #19

Max Range: 0.8592. Min Range: -0.8592

Max Act: 0.6473. Min Act: -0.0000

Data Index: 1653672 (Python Code)

Max Activating Token Index: 145