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

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

Neuron 2027 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.8199. Min Range: -6.8199

Max Act: 6.8199. Min Act: -0.0000

Data Index: 711775 (C4 (Web Text))

Max Activating Token Index: 34

Text #1

Max Range: 6.8199. Min Range: -6.8199

Max Act: 6.0881. Min Act: -0.0001

Data Index: 1193187 (C4 (Web Text))

Max Activating Token Index: 945

Text #2

Max Range: 6.8199. Min Range: -6.8199

Max Act: 6.1250. Min Act: -0.0001

Data Index: 468470 (C4 (Web Text))

Max Activating Token Index: 23

Text #3

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.7581. Min Act: -0.0001

Data Index: 507853 (C4 (Web Text))

Max Activating Token Index: 591

Text #4

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.7229. Min Act: -0.0001

Data Index: 366555 (C4 (Web Text))

Max Activating Token Index: 609

Text #5

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.7371. Min Act: -0.0001

Data Index: 91526 (C4 (Web Text))

Max Activating Token Index: 715

Text #6

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.7316. Min Act: -0.0001

Data Index: 1172821 (C4 (Web Text))

Max Activating Token Index: 227

Text #7

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.6970. Min Act: -0.0001

Data Index: 1047400 (C4 (Web Text))

Max Activating Token Index: 158

Text #8

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.4478. Min Act: -0.0001

Data Index: 395349 (C4 (Web Text))

Max Activating Token Index: 99

Text #9

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.0451. Min Act: -0.0001

Data Index: 508106 (C4 (Web Text))

Max Activating Token Index: 106

Text #10

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.1426. Min Act: -0.0001

Data Index: 666576 (C4 (Web Text))

Max Activating Token Index: 64

Text #11

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.2841. Min Act: -0.0001

Data Index: 827132 (C4 (Web Text))

Max Activating Token Index: 700

Text #12

Max Range: 6.8199. Min Range: -6.8199

Max Act: 3.8056. Min Act: -0.0001

Data Index: 1282187 (C4 (Web Text))

Max Activating Token Index: 678

Text #13

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.1901. Min Act: -0.0001

Data Index: 302043 (C4 (Web Text))

Max Activating Token Index: 244

Text #14

Max Range: 6.8199. Min Range: -6.8199

Max Act: 4.7908. Min Act: -0.0001

Data Index: 743940 (C4 (Web Text))

Max Activating Token Index: 902

Text #15

Max Range: 6.8199. Min Range: -6.8199

Max Act: 5.1508. Min Act: -0.0001

Data Index: 698387 (C4 (Web Text))

Max Activating Token Index: 197

Text #16

Max Range: 6.8199. Min Range: -6.8199

Max Act: 4.7887. Min Act: -0.0000

Data Index: 374992 (C4 (Web Text))

Max Activating Token Index: 36

Text #17

Max Range: 6.8199. Min Range: -6.8199

Max Act: 4.4415. Min Act: -0.0001

Data Index: 377739 (C4 (Web Text))

Max Activating Token Index: 957

Text #18

Max Range: 6.8199. Min Range: -6.8199

Max Act: 4.9232. Min Act: -0.0001

Data Index: 341366 (C4 (Web Text))

Max Activating Token Index: 490

Text #19

Max Range: 6.8199. Min Range: -6.8199

Max Act: 4.8392. Min Act: -0.0001

Data Index: 650057 (C4 (Web Text))

Max Activating Token Index: 24