Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 53 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0394. Min Act: -0.0006

Data Index: 431163 (C4 (Web Text))

Max Activating Token Index: 928

Text #1

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0320. Min Act: -0.0006

Data Index: 552238 (C4 (Web Text))

Max Activating Token Index: 670

Text #2

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0309. Min Act: -0.0006

Data Index: 385206 (C4 (Web Text))

Max Activating Token Index: 846

Text #3

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0361. Min Act: -0.0006

Data Index: 171031 (C4 (Web Text))

Max Activating Token Index: 571

Text #4

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0352. Min Act: -0.0006

Data Index: 1057931 (C4 (Web Text))

Max Activating Token Index: 852

Text #5

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0370. Min Act: -0.0006

Data Index: 829100 (C4 (Web Text))

Max Activating Token Index: 570

Text #6

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0359. Min Act: -0.0006

Data Index: 444428 (C4 (Web Text))

Max Activating Token Index: 815

Text #7

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0392. Min Act: -0.0006

Data Index: 1027054 (C4 (Web Text))

Max Activating Token Index: 973

Text #8

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0303. Min Act: -0.0006

Data Index: 936108 (C4 (Web Text))

Max Activating Token Index: 905

Text #9

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0294. Min Act: -0.0006

Data Index: 168512 (C4 (Web Text))

Max Activating Token Index: 635

Text #10

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0326. Min Act: -0.0006

Data Index: 325677 (C4 (Web Text))

Max Activating Token Index: 282

Text #11

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0368. Min Act: -0.0006

Data Index: 1259120 (C4 (Web Text))

Max Activating Token Index: 900

Text #12

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0333. Min Act: -0.0006

Data Index: 1540049 (Python Code)

Max Activating Token Index: 745

Text #13

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0334. Min Act: -0.0006

Data Index: 471056 (C4 (Web Text))

Max Activating Token Index: 93

Text #14

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0339. Min Act: -0.0006

Data Index: 441152 (C4 (Web Text))

Max Activating Token Index: 638

Text #15

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0286. Min Act: -0.0006

Data Index: 917018 (C4 (Web Text))

Max Activating Token Index: 593

Text #16

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0324. Min Act: -0.0006

Data Index: 794148 (C4 (Web Text))

Max Activating Token Index: 850

Text #17

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0312. Min Act: -0.0006

Data Index: 983648 (C4 (Web Text))

Max Activating Token Index: 583

Text #18

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0311. Min Act: -0.0006

Data Index: 602688 (C4 (Web Text))

Max Activating Token Index: 260

Text #19

Max Range: 0.0394. Min Range: -0.0394

Max Act: 0.0309. Min Act: -0.0006

Data Index: 334863 (C4 (Web Text))

Max Activating Token Index: 748