Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1537 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.9138. Min Act: -0.0001

Data Index: 972661 (C4 (Web Text))

Max Activating Token Index: 98

Text #1

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.6201. Min Act: -0.0001

Data Index: 401002 (C4 (Web Text))

Max Activating Token Index: 39

Text #2

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.5682. Min Act: -0.0001

Data Index: 361878 (C4 (Web Text))

Max Activating Token Index: 36

Text #3

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.4508. Min Act: -0.0001

Data Index: 925453 (C4 (Web Text))

Max Activating Token Index: 73

Text #4

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.4378. Min Act: -0.0001

Data Index: 475876 (C4 (Web Text))

Max Activating Token Index: 439

Text #5

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.3358. Min Act: -0.0001

Data Index: 188927 (C4 (Web Text))

Max Activating Token Index: 126

Text #6

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.0899. Min Act: -0.0001

Data Index: 530901 (C4 (Web Text))

Max Activating Token Index: 31

Text #7

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.8299. Min Act: -0.0001

Data Index: 969885 (C4 (Web Text))

Max Activating Token Index: 59

Text #8

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.8966. Min Act: -0.0001

Data Index: 151895 (C4 (Web Text))

Max Activating Token Index: 356

Text #9

Max Range: 3.9138. Min Range: -3.9138

Max Act: 3.0357. Min Act: -0.0001

Data Index: 699020 (C4 (Web Text))

Max Activating Token Index: 429

Text #10

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.8453. Min Act: -0.0000

Data Index: 470993 (C4 (Web Text))

Max Activating Token Index: 106

Text #11

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.7014. Min Act: -0.0001

Data Index: 1198383 (C4 (Web Text))

Max Activating Token Index: 860

Text #12

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.8027. Min Act: -0.0001

Data Index: 33057 (C4 (Web Text))

Max Activating Token Index: 397

Text #13

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.7831. Min Act: -0.0001

Data Index: 406304 (C4 (Web Text))

Max Activating Token Index: 830

Text #14

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.8665. Min Act: 0.0000

Data Index: 188316 (C4 (Web Text))

Max Activating Token Index: 31

Text #15

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.7207. Min Act: -0.0001

Data Index: 1093402 (C4 (Web Text))

Max Activating Token Index: 108

Text #16

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.7285. Min Act: -0.0001

Data Index: 710647 (C4 (Web Text))

Max Activating Token Index: 160

Text #17

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.5828. Min Act: -0.0001

Data Index: 999744 (C4 (Web Text))

Max Activating Token Index: 54

Text #18

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.5869. Min Act: -0.0001

Data Index: 1311 (C4 (Web Text))

Max Activating Token Index: 225

Text #19

Max Range: 3.9138. Min Range: -3.9138

Max Act: 2.6805. Min Act: -0.0001

Data Index: 991783 (C4 (Web Text))

Max Activating Token Index: 512