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

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

Neuron 4921 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.9451. Min Act: -0.0002

Data Index: 504267 (C4 (Web Text))

Max Activating Token Index: 623

Text #1

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8698. Min Act: -0.0002

Data Index: 901583 (C4 (Web Text))

Max Activating Token Index: 1022

Text #2

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.9443. Min Act: -0.0002

Data Index: 773247 (C4 (Web Text))

Max Activating Token Index: 115

Text #3

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.9048. Min Act: -0.0002

Data Index: 423374 (C4 (Web Text))

Max Activating Token Index: 496

Text #4

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8695. Min Act: -0.0002

Data Index: 1282685 (C4 (Web Text))

Max Activating Token Index: 479

Text #5

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8616. Min Act: -0.0002

Data Index: 1251217 (C4 (Web Text))

Max Activating Token Index: 87

Text #6

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8870. Min Act: -0.0002

Data Index: 1123726 (C4 (Web Text))

Max Activating Token Index: 853

Text #7

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8326. Min Act: -0.0002

Data Index: 1020642 (C4 (Web Text))

Max Activating Token Index: 118

Text #8

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8291. Min Act: -0.0002

Data Index: 706414 (C4 (Web Text))

Max Activating Token Index: 389

Text #9

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8280. Min Act: -0.0002

Data Index: 1070659 (C4 (Web Text))

Max Activating Token Index: 454

Text #10

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8989. Min Act: -0.0002

Data Index: 388931 (C4 (Web Text))

Max Activating Token Index: 890

Text #11

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8731. Min Act: -0.0002

Data Index: 1319670 (C4 (Web Text))

Max Activating Token Index: 741

Text #12

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8083. Min Act: -0.0002

Data Index: 318109 (C4 (Web Text))

Max Activating Token Index: 480

Text #13

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.7869. Min Act: -0.0002

Data Index: 1042248 (C4 (Web Text))

Max Activating Token Index: 640

Text #14

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.9417. Min Act: -0.0002

Data Index: 1222597 (C4 (Web Text))

Max Activating Token Index: 264

Text #15

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8011. Min Act: -0.0002

Data Index: 394482 (C4 (Web Text))

Max Activating Token Index: 697

Text #16

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8115. Min Act: -0.0002

Data Index: 690239 (C4 (Web Text))

Max Activating Token Index: 129

Text #17

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.7689. Min Act: -0.0002

Data Index: 777981 (C4 (Web Text))

Max Activating Token Index: 275

Text #18

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8207. Min Act: -0.0002

Data Index: 855478 (C4 (Web Text))

Max Activating Token Index: 990

Text #19

Max Range: 0.9451. Min Range: -0.9451

Max Act: 0.8140. Min Act: -0.0002

Data Index: 811885 (C4 (Web Text))

Max Activating Token Index: 502