Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 114 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.5986. Min Act: -0.0001

Data Index: 649870 (C4 (Web Text))

Max Activating Token Index: 127

Text #1

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.8640. Min Act: -0.0000

Data Index: 575204 (C4 (Web Text))

Max Activating Token Index: 852

Text #2

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.8209. Min Act: -0.0000

Data Index: 731998 (C4 (Web Text))

Max Activating Token Index: 199

Text #3

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.5059. Min Act: -0.0000

Data Index: 1149368 (C4 (Web Text))

Max Activating Token Index: 348

Text #4

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.4072. Min Act: -0.0001

Data Index: 785174 (C4 (Web Text))

Max Activating Token Index: 584

Text #5

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.3503. Min Act: -0.0000

Data Index: 55863 (C4 (Web Text))

Max Activating Token Index: 353

Text #6

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.1602. Min Act: -0.0001

Data Index: 142379 (C4 (Web Text))

Max Activating Token Index: 54

Text #7

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.3755. Min Act: -0.0000

Data Index: 449614 (C4 (Web Text))

Max Activating Token Index: 670

Text #8

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.8215. Min Act: -0.0000

Data Index: 749318 (C4 (Web Text))

Max Activating Token Index: 654

Text #9

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.9396. Min Act: -0.0000

Data Index: 1007240 (C4 (Web Text))

Max Activating Token Index: 283

Text #10

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.9295. Min Act: -0.0000

Data Index: 12637 (C4 (Web Text))

Max Activating Token Index: 829

Text #11

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.7494. Min Act: -0.0000

Data Index: 739893 (C4 (Web Text))

Max Activating Token Index: 389

Text #12

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.8762. Min Act: -0.0001

Data Index: 280363 (C4 (Web Text))

Max Activating Token Index: 69

Text #13

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.3549. Min Act: -0.0001

Data Index: 768382 (C4 (Web Text))

Max Activating Token Index: 667

Text #14

Max Range: 4.8640. Min Range: -4.8640

Max Act: 4.0272. Min Act: -0.0000

Data Index: 434798 (C4 (Web Text))

Max Activating Token Index: 267

Text #15

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.8374. Min Act: -0.0000

Data Index: 963951 (C4 (Web Text))

Max Activating Token Index: 413

Text #16

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.6801. Min Act: -0.0000

Data Index: 376293 (C4 (Web Text))

Max Activating Token Index: 341

Text #17

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.8163. Min Act: -0.0000

Data Index: 981804 (C4 (Web Text))

Max Activating Token Index: 166

Text #18

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.7663. Min Act: -0.0000

Data Index: 942389 (C4 (Web Text))

Max Activating Token Index: 338

Text #19

Max Range: 4.8640. Min Range: -4.8640

Max Act: 3.9234. Min Act: -0.0001

Data Index: 678362 (C4 (Web Text))

Max Activating Token Index: 99