Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1009 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1502. Min Act: -0.0003

Data Index: 195187 (C4 (Web Text))

Max Activating Token Index: 731

Text #1

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1073. Min Act: -0.0003

Data Index: 1978 (C4 (Web Text))

Max Activating Token Index: 957

Text #2

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1075. Min Act: -0.0003

Data Index: 691024 (C4 (Web Text))

Max Activating Token Index: 172

Text #3

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1088. Min Act: -0.0003

Data Index: 756611 (C4 (Web Text))

Max Activating Token Index: 677

Text #4

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1074. Min Act: -0.0003

Data Index: 1345876 (C4 (Web Text))

Max Activating Token Index: 328

Text #5

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1060. Min Act: -0.0003

Data Index: 759423 (C4 (Web Text))

Max Activating Token Index: 310

Text #6

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1061. Min Act: -0.0003

Data Index: 30612 (C4 (Web Text))

Max Activating Token Index: 20

Text #7

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0992. Min Act: -0.0004

Data Index: 1059189 (C4 (Web Text))

Max Activating Token Index: 630

Text #8

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1023. Min Act: -0.0002

Data Index: 668585 (C4 (Web Text))

Max Activating Token Index: 820

Text #9

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0990. Min Act: -0.0003

Data Index: 1595590 (Python Code)

Max Activating Token Index: 544

Text #10

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1008. Min Act: -0.0003

Data Index: 571517 (C4 (Web Text))

Max Activating Token Index: 926

Text #11

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1002. Min Act: -0.0003

Data Index: 1200145 (C4 (Web Text))

Max Activating Token Index: 379

Text #12

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0948. Min Act: -0.0002

Data Index: 1029527 (C4 (Web Text))

Max Activating Token Index: 559

Text #13

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0984. Min Act: -0.0003

Data Index: 586400 (C4 (Web Text))

Max Activating Token Index: 931

Text #14

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.1016. Min Act: -0.0003

Data Index: 502248 (C4 (Web Text))

Max Activating Token Index: 599

Text #15

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0970. Min Act: -0.0003

Data Index: 871176 (C4 (Web Text))

Max Activating Token Index: 824

Text #16

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0935. Min Act: -0.0003

Data Index: 135156 (C4 (Web Text))

Max Activating Token Index: 39

Text #17

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0969. Min Act: -0.0003

Data Index: 19198 (C4 (Web Text))

Max Activating Token Index: 910

Text #18

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0996. Min Act: -0.0003

Data Index: 1055774 (C4 (Web Text))

Max Activating Token Index: 855

Text #19

Max Range: 0.1502. Min Range: -0.1502

Max Act: 0.0964. Min Act: -0.0003

Data Index: 1326162 (C4 (Web Text))

Max Activating Token Index: 785