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

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

Neuron 4047 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: 0.7781. Min Range: -0.7781

Max Act: 0.7781. Min Act: -0.0001

Data Index: 358469 (C4 (Web Text))

Max Activating Token Index: 627

Text #1

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.6143. Min Act: -0.0001

Data Index: 1116982 (C4 (Web Text))

Max Activating Token Index: 86

Text #2

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.6040. Min Act: -0.0001

Data Index: 212904 (C4 (Web Text))

Max Activating Token Index: 183

Text #3

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5370. Min Act: -0.0001

Data Index: 660513 (C4 (Web Text))

Max Activating Token Index: 792

Text #4

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5969. Min Act: -0.0001

Data Index: 852282 (C4 (Web Text))

Max Activating Token Index: 964

Text #5

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5348. Min Act: -0.0001

Data Index: 513154 (C4 (Web Text))

Max Activating Token Index: 740

Text #6

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5096. Min Act: -0.0001

Data Index: 575367 (C4 (Web Text))

Max Activating Token Index: 26

Text #7

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4983. Min Act: -0.0001

Data Index: 404232 (C4 (Web Text))

Max Activating Token Index: 813

Text #8

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5125. Min Act: -0.0001

Data Index: 975390 (C4 (Web Text))

Max Activating Token Index: 128

Text #9

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5062. Min Act: -0.0001

Data Index: 131323 (C4 (Web Text))

Max Activating Token Index: 431

Text #10

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5187. Min Act: -0.0001

Data Index: 1096423 (C4 (Web Text))

Max Activating Token Index: 283

Text #11

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4189. Min Act: -0.0001

Data Index: 1334521 (C4 (Web Text))

Max Activating Token Index: 151

Text #12

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4610. Min Act: -0.0001

Data Index: 670485 (C4 (Web Text))

Max Activating Token Index: 8

Text #13

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4787. Min Act: -0.0001

Data Index: 596086 (C4 (Web Text))

Max Activating Token Index: 546

Text #14

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4581. Min Act: -0.0001

Data Index: 926360 (C4 (Web Text))

Max Activating Token Index: 60

Text #15

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4712. Min Act: -0.0001

Data Index: 648781 (C4 (Web Text))

Max Activating Token Index: 458

Text #16

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.5098. Min Act: -0.0001

Data Index: 140835 (C4 (Web Text))

Max Activating Token Index: 221

Text #17

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4779. Min Act: -0.0001

Data Index: 1621174 (Python Code)

Max Activating Token Index: 174

Text #18

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4071. Min Act: -0.0001

Data Index: 570046 (C4 (Web Text))

Max Activating Token Index: 961

Text #19

Max Range: 0.7781. Min Range: -0.7781

Max Act: 0.4622. Min Act: -0.0001

Data Index: 766462 (C4 (Web Text))

Max Activating Token Index: 680