Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 961 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.2097. Min Act: -0.0003

Data Index: 593348 (C4 (Web Text))

Max Activating Token Index: 902

Text #1

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1800. Min Act: -0.0003

Data Index: 1235809 (C4 (Web Text))

Max Activating Token Index: 370

Text #2

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1859. Min Act: -0.0003

Data Index: 789393 (C4 (Web Text))

Max Activating Token Index: 797

Text #3

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1802. Min Act: -0.0003

Data Index: 836415 (C4 (Web Text))

Max Activating Token Index: 69

Text #4

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1700. Min Act: -0.0003

Data Index: 58139 (C4 (Web Text))

Max Activating Token Index: 815

Text #5

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1709. Min Act: -0.0003

Data Index: 132250 (C4 (Web Text))

Max Activating Token Index: 216

Text #6

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1508. Min Act: -0.0003

Data Index: 1212180 (C4 (Web Text))

Max Activating Token Index: 719

Text #7

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1494. Min Act: -0.0003

Data Index: 547841 (C4 (Web Text))

Max Activating Token Index: 195

Text #8

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1471. Min Act: -0.0003

Data Index: 1611260 (Python Code)

Max Activating Token Index: 338

Text #9

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1456. Min Act: -0.0003

Data Index: 1473390 (Python Code)

Max Activating Token Index: 972

Text #10

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1503. Min Act: -0.0003

Data Index: 876724 (C4 (Web Text))

Max Activating Token Index: 741

Text #11

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1420. Min Act: -0.0003

Data Index: 1003330 (C4 (Web Text))

Max Activating Token Index: 903

Text #12

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1530. Min Act: -0.0003

Data Index: 1147975 (C4 (Web Text))

Max Activating Token Index: 791

Text #13

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1231. Min Act: -0.0003

Data Index: 29182 (C4 (Web Text))

Max Activating Token Index: 611

Text #14

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1209. Min Act: -0.0003

Data Index: 849782 (C4 (Web Text))

Max Activating Token Index: 670

Text #15

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1440. Min Act: -0.0003

Data Index: 1458344 (Python Code)

Max Activating Token Index: 315

Text #16

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1405. Min Act: -0.0003

Data Index: 372252 (C4 (Web Text))

Max Activating Token Index: 369

Text #17

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1499. Min Act: -0.0003

Data Index: 1212480 (C4 (Web Text))

Max Activating Token Index: 391

Text #18

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1430. Min Act: -0.0003

Data Index: 921051 (C4 (Web Text))

Max Activating Token Index: 763

Text #19

Max Range: 0.2097. Min Range: -0.2097

Max Act: 0.1412. Min Act: -0.0003

Data Index: 1013463 (C4 (Web Text))

Max Activating Token Index: 249