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

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

Neuron 4530 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: 3.6987. Min Range: -3.6987

Max Act: 3.6987. Min Act: -0.0000

Data Index: 937185 (C4 (Web Text))

Max Activating Token Index: 478

Text #1

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.5123. Min Act: -0.0000

Data Index: 868500 (C4 (Web Text))

Max Activating Token Index: 476

Text #2

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3477. Min Act: -0.0000

Data Index: 1006248 (C4 (Web Text))

Max Activating Token Index: 688

Text #3

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2204. Min Act: -0.0000

Data Index: 871007 (C4 (Web Text))

Max Activating Token Index: 259

Text #4

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.3682. Min Act: -0.0000

Data Index: 1587832 (Python Code)

Max Activating Token Index: 970

Text #5

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.1995. Min Act: -0.0000

Data Index: 1466570 (Python Code)

Max Activating Token Index: 828

Text #6

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.1338. Min Act: -0.0000

Data Index: 1591916 (Python Code)

Max Activating Token Index: 1003

Text #7

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.2359. Min Act: -0.0000

Data Index: 1509985 (Python Code)

Max Activating Token Index: 961

Text #8

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.0114. Min Act: -0.0000

Data Index: 1581827 (Python Code)

Max Activating Token Index: 729

Text #9

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.0195. Min Act: -0.0000

Data Index: 1194241 (C4 (Web Text))

Max Activating Token Index: 904

Text #10

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.0053. Min Act: -0.0000

Data Index: 1430229 (Python Code)

Max Activating Token Index: 912

Text #11

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.8690. Min Act: -0.0000

Data Index: 1634550 (Python Code)

Max Activating Token Index: 1012

Text #12

Max Range: 3.6987. Min Range: -3.6987

Max Act: 3.1570. Min Act: -0.0000

Data Index: 612430 (C4 (Web Text))

Max Activating Token Index: 940

Text #13

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.8606. Min Act: -0.0000

Data Index: 1503067 (Python Code)

Max Activating Token Index: 748

Text #14

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.9588. Min Act: -0.0000

Data Index: 1335121 (C4 (Web Text))

Max Activating Token Index: 230

Text #15

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.7857. Min Act: -0.0000

Data Index: 103150 (C4 (Web Text))

Max Activating Token Index: 350

Text #16

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.9311. Min Act: -0.0000

Data Index: 1621709 (Python Code)

Max Activating Token Index: 102

Text #17

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.8347. Min Act: -0.0000

Data Index: 147378 (C4 (Web Text))

Max Activating Token Index: 717

Text #18

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.8003. Min Act: -0.0000

Data Index: 1225905 (C4 (Web Text))

Max Activating Token Index: 248

Text #19

Max Range: 3.6987. Min Range: -3.6987

Max Act: 2.6279. Min Act: -0.0000

Data Index: 989689 (C4 (Web Text))

Max Activating Token Index: 859