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

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

Neuron 5017 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: 0.8689. Min Range: -0.8689

Max Act: 0.8218. Min Act: -0.0000

Data Index: 1453114 (Python Code)

Max Activating Token Index: 219

Text #1

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.8689. Min Act: -0.0000

Data Index: 319742 (C4 (Web Text))

Max Activating Token Index: 40

Text #2

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7893. Min Act: -0.0000

Data Index: 1207885 (C4 (Web Text))

Max Activating Token Index: 320

Text #3

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.8254. Min Act: -0.0000

Data Index: 870128 (C4 (Web Text))

Max Activating Token Index: 430

Text #4

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.8214. Min Act: -0.0000

Data Index: 1081219 (C4 (Web Text))

Max Activating Token Index: 595

Text #5

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7641. Min Act: -0.0000

Data Index: 193502 (C4 (Web Text))

Max Activating Token Index: 343

Text #6

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7855. Min Act: -0.0000

Data Index: 967670 (C4 (Web Text))

Max Activating Token Index: 1002

Text #7

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7781. Min Act: -0.0000

Data Index: 1316082 (C4 (Web Text))

Max Activating Token Index: 38

Text #8

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7265. Min Act: -0.0000

Data Index: 111898 (C4 (Web Text))

Max Activating Token Index: 638

Text #9

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7360. Min Act: -0.0000

Data Index: 1316698 (C4 (Web Text))

Max Activating Token Index: 838

Text #10

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7605. Min Act: -0.0000

Data Index: 1301700 (C4 (Web Text))

Max Activating Token Index: 443

Text #11

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.6862. Min Act: -0.0000

Data Index: 462405 (C4 (Web Text))

Max Activating Token Index: 216

Text #12

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7056. Min Act: -0.0000

Data Index: 315159 (C4 (Web Text))

Max Activating Token Index: 936

Text #13

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7030. Min Act: -0.0000

Data Index: 1629020 (Python Code)

Max Activating Token Index: 58

Text #14

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7232. Min Act: -0.0000

Data Index: 1349218 (C4 (Web Text))

Max Activating Token Index: 294

Text #15

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7431. Min Act: -0.0000

Data Index: 112997 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.6769. Min Act: -0.0000

Data Index: 1346299 (C4 (Web Text))

Max Activating Token Index: 808

Text #17

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.6382. Min Act: -0.0000

Data Index: 419725 (C4 (Web Text))

Max Activating Token Index: 988

Text #18

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.6795. Min Act: -0.0000

Data Index: 183292 (C4 (Web Text))

Max Activating Token Index: 390

Text #19

Max Range: 0.8689. Min Range: -0.8689

Max Act: 0.7036. Min Act: -0.0000

Data Index: 638730 (C4 (Web Text))

Max Activating Token Index: 227