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

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

Neuron 5034 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1216. Min Range: -1.1216

Max Act: 1.1216. Min Act: -0.0000

Data Index: 783851 (C4 (Web Text))

Max Activating Token Index: 29

Text #1

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9562. Min Act: -0.0000

Data Index: 349986 (C4 (Web Text))

Max Activating Token Index: 443

Text #2

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8881. Min Act: -0.0000

Data Index: 1236285 (C4 (Web Text))

Max Activating Token Index: 945

Text #3

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9909. Min Act: -0.0000

Data Index: 1216722 (C4 (Web Text))

Max Activating Token Index: 973

Text #4

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9048. Min Act: -0.0000

Data Index: 207556 (C4 (Web Text))

Max Activating Token Index: 869

Text #5

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9058. Min Act: -0.0000

Data Index: 921172 (C4 (Web Text))

Max Activating Token Index: 32

Text #6

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9387. Min Act: -0.0000

Data Index: 1126115 (C4 (Web Text))

Max Activating Token Index: 32

Text #7

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9512. Min Act: -0.0000

Data Index: 215157 (C4 (Web Text))

Max Activating Token Index: 121

Text #8

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9345. Min Act: -0.0000

Data Index: 1143104 (C4 (Web Text))

Max Activating Token Index: 1001

Text #9

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8810. Min Act: -0.0000

Data Index: 1185788 (C4 (Web Text))

Max Activating Token Index: 48

Text #10

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8938. Min Act: -0.0000

Data Index: 402475 (C4 (Web Text))

Max Activating Token Index: 83

Text #11

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8870. Min Act: -0.0000

Data Index: 65034 (C4 (Web Text))

Max Activating Token Index: 847

Text #12

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9149. Min Act: -0.0000

Data Index: 1579697 (Python Code)

Max Activating Token Index: 133

Text #13

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.9443. Min Act: -0.0000

Data Index: 1350388 (C4 (Web Text))

Max Activating Token Index: 87

Text #14

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8512. Min Act: -0.0000

Data Index: 1003252 (C4 (Web Text))

Max Activating Token Index: 118

Text #15

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.7857. Min Act: -0.0000

Data Index: 146426 (C4 (Web Text))

Max Activating Token Index: 94

Text #16

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8645. Min Act: -0.0000

Data Index: 579596 (C4 (Web Text))

Max Activating Token Index: 480

Text #17

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8036. Min Act: -0.0000

Data Index: 656156 (C4 (Web Text))

Max Activating Token Index: 419

Text #18

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.8880. Min Act: -0.0000

Data Index: 1074350 (C4 (Web Text))

Max Activating Token Index: 102

Text #19

Max Range: 1.1216. Min Range: -1.1216

Max Act: 0.7535. Min Act: -0.0000

Data Index: 1399010 (Python Code)

Max Activating Token Index: 677