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

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

Neuron 3862 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8609. Min Range: -5.8609

Max Act: 5.7510. Min Act: -0.0000

Data Index: 44342 (C4 (Web Text))

Max Activating Token Index: 413

Text #1

Max Range: 5.8609. Min Range: -5.8609

Max Act: 5.2726. Min Act: -0.0000

Data Index: 797284 (C4 (Web Text))

Max Activating Token Index: 707

Text #2

Max Range: 5.8609. Min Range: -5.8609

Max Act: 5.8609. Min Act: -0.0000

Data Index: 805170 (C4 (Web Text))

Max Activating Token Index: 215

Text #3

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.8345. Min Act: -0.0000

Data Index: 887779 (C4 (Web Text))

Max Activating Token Index: 327

Text #4

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.5611. Min Act: -0.0000

Data Index: 107549 (C4 (Web Text))

Max Activating Token Index: 928

Text #5

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.2632. Min Act: -0.0000

Data Index: 1012101 (C4 (Web Text))

Max Activating Token Index: 656

Text #6

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.2936. Min Act: -0.0000

Data Index: 1175926 (C4 (Web Text))

Max Activating Token Index: 961

Text #7

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.3617. Min Act: -0.0000

Data Index: 497067 (C4 (Web Text))

Max Activating Token Index: 61

Text #8

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.0127. Min Act: -0.0000

Data Index: 19063 (C4 (Web Text))

Max Activating Token Index: 1021

Text #9

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.9969. Min Act: -0.0000

Data Index: 32052 (C4 (Web Text))

Max Activating Token Index: 76

Text #10

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.0875. Min Act: -0.0000

Data Index: 331690 (C4 (Web Text))

Max Activating Token Index: 963

Text #11

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.7335. Min Act: -0.0000

Data Index: 202968 (C4 (Web Text))

Max Activating Token Index: 485

Text #12

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.0425. Min Act: -0.0000

Data Index: 1163528 (C4 (Web Text))

Max Activating Token Index: 939

Text #13

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.9119. Min Act: -0.0000

Data Index: 150870 (C4 (Web Text))

Max Activating Token Index: 807

Text #14

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.6262. Min Act: -0.0000

Data Index: 101531 (C4 (Web Text))

Max Activating Token Index: 326

Text #15

Max Range: 5.8609. Min Range: -5.8609

Max Act: 4.0987. Min Act: -0.0000

Data Index: 1041541 (C4 (Web Text))

Max Activating Token Index: 864

Text #16

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.7404. Min Act: -0.0000

Data Index: 1643434 (Python Code)

Max Activating Token Index: 513

Text #17

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.7552. Min Act: -0.0000

Data Index: 194854 (C4 (Web Text))

Max Activating Token Index: 114

Text #18

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.1676. Min Act: -0.0000

Data Index: 534165 (C4 (Web Text))

Max Activating Token Index: 804

Text #19

Max Range: 5.8609. Min Range: -5.8609

Max Act: 3.3550. Min Act: -0.0000

Data Index: 622215 (C4 (Web Text))

Max Activating Token Index: 904