Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3299 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 3.0002. Min Range: -3.0002

Max Act: 3.0002. Min Act: -0.0001

Data Index: 581716 (C4 (Web Text))

Max Activating Token Index: 287

Text #1

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.8725. Min Act: -0.0001

Data Index: 1298379 (C4 (Web Text))

Max Activating Token Index: 809

Text #2

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.6343. Min Act: -0.0001

Data Index: 295666 (C4 (Web Text))

Max Activating Token Index: 499

Text #3

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.4821. Min Act: -0.0001

Data Index: 900679 (C4 (Web Text))

Max Activating Token Index: 860

Text #4

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.6242. Min Act: -0.0001

Data Index: 870228 (C4 (Web Text))

Max Activating Token Index: 916

Text #5

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.8170. Min Act: -0.0001

Data Index: 908409 (C4 (Web Text))

Max Activating Token Index: 241

Text #6

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.7096. Min Act: -0.0001

Data Index: 1067239 (C4 (Web Text))

Max Activating Token Index: 273

Text #7

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.2518. Min Act: -0.0001

Data Index: 953085 (C4 (Web Text))

Max Activating Token Index: 972

Text #8

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.4583. Min Act: -0.0001

Data Index: 387019 (C4 (Web Text))

Max Activating Token Index: 1007

Text #9

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.8062. Min Act: -0.0001

Data Index: 348517 (C4 (Web Text))

Max Activating Token Index: 588

Text #10

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.3796. Min Act: -0.0001

Data Index: 734108 (C4 (Web Text))

Max Activating Token Index: 711

Text #11

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.1230. Min Act: -0.0001

Data Index: 448623 (C4 (Web Text))

Max Activating Token Index: 817

Text #12

Max Range: 3.0002. Min Range: -3.0002

Max Act: 1.9768. Min Act: -0.0001

Data Index: 774617 (C4 (Web Text))

Max Activating Token Index: 241

Text #13

Max Range: 3.0002. Min Range: -3.0002

Max Act: 1.9149. Min Act: -0.0001

Data Index: 34467 (C4 (Web Text))

Max Activating Token Index: 135

Text #14

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.2568. Min Act: -0.0001

Data Index: 1282591 (C4 (Web Text))

Max Activating Token Index: 860

Text #15

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.2320. Min Act: -0.0001

Data Index: 1045560 (C4 (Web Text))

Max Activating Token Index: 568

Text #16

Max Range: 3.0002. Min Range: -3.0002

Max Act: 1.9741. Min Act: -0.0001

Data Index: 898542 (C4 (Web Text))

Max Activating Token Index: 138

Text #17

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.3347. Min Act: -0.0001

Data Index: 642782 (C4 (Web Text))

Max Activating Token Index: 90

Text #18

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.0766. Min Act: -0.0001

Data Index: 1203736 (C4 (Web Text))

Max Activating Token Index: 743

Text #19

Max Range: 3.0002. Min Range: -3.0002

Max Act: 2.0332. Min Act: -0.0001

Data Index: 944592 (C4 (Web Text))

Max Activating Token Index: 265