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

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

Neuron 3173 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.4678. Min Act: -0.0001

Data Index: 597932 (C4 (Web Text))

Max Activating Token Index: 1017

Text #1

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.4358. Min Act: -0.0001

Data Index: 1454799 (Python Code)

Max Activating Token Index: 137

Text #2

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.4220. Min Act: -0.0000

Data Index: 1301219 (C4 (Web Text))

Max Activating Token Index: 252

Text #3

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.4112. Min Act: -0.0001

Data Index: 1476285 (Python Code)

Max Activating Token Index: 85

Text #4

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.4242. Min Act: -0.0001

Data Index: 1426575 (Python Code)

Max Activating Token Index: 543

Text #5

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3961. Min Act: -0.0001

Data Index: 1583444 (Python Code)

Max Activating Token Index: 143

Text #6

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3443. Min Act: -0.0000

Data Index: 1064359 (C4 (Web Text))

Max Activating Token Index: 346

Text #7

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3823. Min Act: -0.0001

Data Index: 1590208 (Python Code)

Max Activating Token Index: 626

Text #8

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3782. Min Act: -0.0000

Data Index: 883162 (C4 (Web Text))

Max Activating Token Index: 544

Text #9

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.4157. Min Act: -0.0001

Data Index: 1633720 (Python Code)

Max Activating Token Index: 80

Text #10

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3677. Min Act: -0.0000

Data Index: 1204497 (C4 (Web Text))

Max Activating Token Index: 829

Text #11

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3835. Min Act: -0.0000

Data Index: 584608 (C4 (Web Text))

Max Activating Token Index: 308

Text #12

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3971. Min Act: -0.0000

Data Index: 267995 (C4 (Web Text))

Max Activating Token Index: 707

Text #13

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3671. Min Act: -0.0001

Data Index: 1651149 (Python Code)

Max Activating Token Index: 70

Text #14

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3701. Min Act: -0.0001

Data Index: 1641466 (Python Code)

Max Activating Token Index: 157

Text #15

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3617. Min Act: -0.0001

Data Index: 1569374 (Python Code)

Max Activating Token Index: 753

Text #16

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3798. Min Act: -0.0000

Data Index: 1085766 (C4 (Web Text))

Max Activating Token Index: 913

Text #17

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3583. Min Act: -0.0000

Data Index: 135923 (C4 (Web Text))

Max Activating Token Index: 731

Text #18

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3698. Min Act: -0.0001

Data Index: 1049028 (C4 (Web Text))

Max Activating Token Index: 1011

Text #19

Max Range: 0.4678. Min Range: -0.4678

Max Act: 0.3899. Min Act: -0.0001

Data Index: 1451416 (Python Code)

Max Activating Token Index: 183