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

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

Neuron 3952 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: 0.5432. Min Range: -0.5432

Max Act: 0.5049. Min Act: -0.0000

Data Index: 1210784 (C4 (Web Text))

Max Activating Token Index: 701

Text #1

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.5432. Min Act: -0.0000

Data Index: 1220354 (C4 (Web Text))

Max Activating Token Index: 944

Text #2

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4815. Min Act: -0.0000

Data Index: 737111 (C4 (Web Text))

Max Activating Token Index: 671

Text #3

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.5314. Min Act: -0.0000

Data Index: 415085 (C4 (Web Text))

Max Activating Token Index: 322

Text #4

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4930. Min Act: -0.0000

Data Index: 397629 (C4 (Web Text))

Max Activating Token Index: 377

Text #5

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.5024. Min Act: -0.0000

Data Index: 291715 (C4 (Web Text))

Max Activating Token Index: 581

Text #6

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4504. Min Act: -0.0000

Data Index: 492026 (C4 (Web Text))

Max Activating Token Index: 503

Text #7

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4633. Min Act: -0.0000

Data Index: 839431 (C4 (Web Text))

Max Activating Token Index: 422

Text #8

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.5021. Min Act: -0.0000

Data Index: 754006 (C4 (Web Text))

Max Activating Token Index: 279

Text #9

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4534. Min Act: -0.0000

Data Index: 1160837 (C4 (Web Text))

Max Activating Token Index: 785

Text #10

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4300. Min Act: -0.0000

Data Index: 863483 (C4 (Web Text))

Max Activating Token Index: 424

Text #11

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4418. Min Act: -0.0000

Data Index: 503558 (C4 (Web Text))

Max Activating Token Index: 248

Text #12

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4671. Min Act: -0.0000

Data Index: 669005 (C4 (Web Text))

Max Activating Token Index: 819

Text #13

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4320. Min Act: -0.0000

Data Index: 124752 (C4 (Web Text))

Max Activating Token Index: 255

Text #14

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.4228. Min Act: -0.0000

Data Index: 435682 (C4 (Web Text))

Max Activating Token Index: 795

Text #15

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.3785. Min Act: -0.0000

Data Index: 1529009 (Python Code)

Max Activating Token Index: 753

Text #16

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.3672. Min Act: -0.0000

Data Index: 122492 (C4 (Web Text))

Max Activating Token Index: 166

Text #17

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.2986. Min Act: -0.0000

Data Index: 91980 (C4 (Web Text))

Max Activating Token Index: 574

Text #18

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.3291. Min Act: -0.0000

Data Index: 1566987 (Python Code)

Max Activating Token Index: 84

Text #19

Max Range: 0.5432. Min Range: -0.5432

Max Act: 0.3239. Min Act: -0.0000

Data Index: 752349 (C4 (Web Text))

Max Activating Token Index: 415