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

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

Neuron 4458 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.9000. Min Act: -0.0000

Data Index: 99218 (C4 (Web Text))

Max Activating Token Index: 314

Text #1

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.6330. Min Act: 0.0000

Data Index: 152654 (C4 (Web Text))

Max Activating Token Index: 410

Text #2

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.6286. Min Act: -0.0000

Data Index: 33335 (C4 (Web Text))

Max Activating Token Index: 563

Text #3

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.6212. Min Act: -0.0000

Data Index: 91823 (C4 (Web Text))

Max Activating Token Index: 963

Text #4

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.6275. Min Act: -0.0000

Data Index: 939906 (C4 (Web Text))

Max Activating Token Index: 845

Text #5

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5625. Min Act: 0.0000

Data Index: 230069 (C4 (Web Text))

Max Activating Token Index: 955

Text #6

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5837. Min Act: -0.0000

Data Index: 450174 (C4 (Web Text))

Max Activating Token Index: 851

Text #7

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5812. Min Act: -0.0000

Data Index: 1074383 (C4 (Web Text))

Max Activating Token Index: 93

Text #8

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5557. Min Act: -0.0000

Data Index: 288888 (C4 (Web Text))

Max Activating Token Index: 719

Text #9

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5621. Min Act: -0.0000

Data Index: 15180 (C4 (Web Text))

Max Activating Token Index: 660

Text #10

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5354. Min Act: -0.0000

Data Index: 748989 (C4 (Web Text))

Max Activating Token Index: 744

Text #11

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5160. Min Act: -0.0000

Data Index: 675908 (C4 (Web Text))

Max Activating Token Index: 277

Text #12

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5045. Min Act: -0.0000

Data Index: 70092 (C4 (Web Text))

Max Activating Token Index: 303

Text #13

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5044. Min Act: -0.0000

Data Index: 991714 (C4 (Web Text))

Max Activating Token Index: 244

Text #14

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.4943. Min Act: -0.0000

Data Index: 231000 (C4 (Web Text))

Max Activating Token Index: 424

Text #15

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5091. Min Act: -0.0000

Data Index: 1467510 (Python Code)

Max Activating Token Index: 920

Text #16

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5058. Min Act: -0.0000

Data Index: 219206 (C4 (Web Text))

Max Activating Token Index: 986

Text #17

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5045. Min Act: -0.0000

Data Index: 1187091 (C4 (Web Text))

Max Activating Token Index: 346

Text #18

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.5000. Min Act: -0.0000

Data Index: 1071363 (C4 (Web Text))

Max Activating Token Index: 905

Text #19

Max Range: 0.9000. Min Range: -0.9000

Max Act: 0.4767. Min Act: 0.0001

Data Index: 987635 (C4 (Web Text))

Max Activating Token Index: 690