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

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

Neuron 1912 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.8880. Min Act: -0.0000

Data Index: 879447 (C4 (Web Text))

Max Activating Token Index: 858

Text #1

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.2003. Min Act: -0.0000

Data Index: 328739 (C4 (Web Text))

Max Activating Token Index: 918

Text #2

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.0693. Min Act: -0.0000

Data Index: 248980 (C4 (Web Text))

Max Activating Token Index: 790

Text #3

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.1920. Min Act: -0.0000

Data Index: 161225 (C4 (Web Text))

Max Activating Token Index: 651

Text #4

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.1955. Min Act: -0.0000

Data Index: 63031 (C4 (Web Text))

Max Activating Token Index: 653

Text #5

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.2487. Min Act: -0.0000

Data Index: 92886 (C4 (Web Text))

Max Activating Token Index: 924

Text #6

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.1998. Min Act: -0.0000

Data Index: 1147500 (C4 (Web Text))

Max Activating Token Index: 696

Text #7

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.1669. Min Act: -0.0000

Data Index: 258836 (C4 (Web Text))

Max Activating Token Index: 835

Text #8

Max Range: 4.2487. Min Range: -4.2487

Max Act: 4.1604. Min Act: -0.0000

Data Index: 1321660 (C4 (Web Text))

Max Activating Token Index: 877

Text #9

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.8115. Min Act: -0.0000

Data Index: 1445190 (Python Code)

Max Activating Token Index: 429

Text #10

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.8067. Min Act: -0.0000

Data Index: 90362 (C4 (Web Text))

Max Activating Token Index: 4

Text #11

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.9719. Min Act: -0.0000

Data Index: 1656601 (Python Code)

Max Activating Token Index: 403

Text #12

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.8574. Min Act: -0.0000

Data Index: 895687 (C4 (Web Text))

Max Activating Token Index: 812

Text #13

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.8323. Min Act: -0.0000

Data Index: 1326913 (C4 (Web Text))

Max Activating Token Index: 200

Text #14

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.6490. Min Act: -0.0000

Data Index: 1219040 (C4 (Web Text))

Max Activating Token Index: 70

Text #15

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.8009. Min Act: -0.0000

Data Index: 1450011 (Python Code)

Max Activating Token Index: 536

Text #16

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.7340. Min Act: -0.0000

Data Index: 409715 (C4 (Web Text))

Max Activating Token Index: 335

Text #17

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.6892. Min Act: -0.0000

Data Index: 252709 (C4 (Web Text))

Max Activating Token Index: 953

Text #18

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.6682. Min Act: -0.0000

Data Index: 504416 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 4.2487. Min Range: -4.2487

Max Act: 3.4686. Min Act: -0.0000

Data Index: 1026941 (C4 (Web Text))

Max Activating Token Index: 20