Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1538 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.6300. Min Act: -0.0000

Data Index: 180077 (C4 (Web Text))

Max Activating Token Index: 559

Text #1

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.5451. Min Act: -0.0000

Data Index: 1281911 (C4 (Web Text))

Max Activating Token Index: 24

Text #2

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.3230. Min Act: -0.0000

Data Index: 693997 (C4 (Web Text))

Max Activating Token Index: 710

Text #3

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.2091. Min Act: -0.0000

Data Index: 1279559 (C4 (Web Text))

Max Activating Token Index: 1011

Text #4

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.1849. Min Act: -0.0000

Data Index: 14461 (C4 (Web Text))

Max Activating Token Index: 925

Text #5

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.1374. Min Act: -0.0000

Data Index: 409120 (C4 (Web Text))

Max Activating Token Index: 494

Text #6

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.0705. Min Act: -0.0000

Data Index: 316492 (C4 (Web Text))

Max Activating Token Index: 351

Text #7

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.1718. Min Act: -0.0000

Data Index: 400523 (C4 (Web Text))

Max Activating Token Index: 275

Text #8

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.0493. Min Act: -0.0000

Data Index: 708627 (C4 (Web Text))

Max Activating Token Index: 690

Text #9

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.1223. Min Act: -0.0000

Data Index: 495599 (C4 (Web Text))

Max Activating Token Index: 704

Text #10

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.0129. Min Act: -0.0000

Data Index: 90275 (C4 (Web Text))

Max Activating Token Index: 375

Text #11

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.8975. Min Act: -0.0000

Data Index: 408162 (C4 (Web Text))

Max Activating Token Index: 663

Text #12

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.0040. Min Act: -0.0000

Data Index: 162687 (C4 (Web Text))

Max Activating Token Index: 66

Text #13

Max Range: 2.6300. Min Range: -2.6300

Max Act: 2.0112. Min Act: -0.0000

Data Index: 1509116 (Python Code)

Max Activating Token Index: 800

Text #14

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.8694. Min Act: -0.0000

Data Index: 1123313 (C4 (Web Text))

Max Activating Token Index: 767

Text #15

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.9447. Min Act: -0.0000

Data Index: 807233 (C4 (Web Text))

Max Activating Token Index: 924

Text #16

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.9015. Min Act: -0.0000

Data Index: 60662 (C4 (Web Text))

Max Activating Token Index: 233

Text #17

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.9397. Min Act: -0.0000

Data Index: 1019441 (C4 (Web Text))

Max Activating Token Index: 662

Text #18

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.9356. Min Act: -0.0000

Data Index: 1125972 (C4 (Web Text))

Max Activating Token Index: 194

Text #19

Max Range: 2.6300. Min Range: -2.6300

Max Act: 1.9824. Min Act: -0.0000

Data Index: 228195 (C4 (Web Text))

Max Activating Token Index: 990