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

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

Neuron 1969 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.4188. Min Act: -0.0000

Data Index: 897932 (C4 (Web Text))

Max Activating Token Index: 943

Text #1

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.3054. Min Act: -0.0001

Data Index: 768110 (C4 (Web Text))

Max Activating Token Index: 479

Text #2

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2732. Min Act: -0.0000

Data Index: 729057 (C4 (Web Text))

Max Activating Token Index: 664

Text #3

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2571. Min Act: -0.0000

Data Index: 464461 (C4 (Web Text))

Max Activating Token Index: 91

Text #4

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2462. Min Act: -0.0001

Data Index: 318141 (C4 (Web Text))

Max Activating Token Index: 187

Text #5

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2438. Min Act: -0.0000

Data Index: 699556 (C4 (Web Text))

Max Activating Token Index: 1003

Text #6

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2210. Min Act: -0.0000

Data Index: 1134261 (C4 (Web Text))

Max Activating Token Index: 790

Text #7

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2136. Min Act: -0.0001

Data Index: 239489 (C4 (Web Text))

Max Activating Token Index: 283

Text #8

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2138. Min Act: -0.0001

Data Index: 1369883 (Python Code)

Max Activating Token Index: 34

Text #9

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2120. Min Act: -0.0000

Data Index: 123900 (C4 (Web Text))

Max Activating Token Index: 541

Text #10

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2006. Min Act: -0.0001

Data Index: 937146 (C4 (Web Text))

Max Activating Token Index: 905

Text #11

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2015. Min Act: -0.0000

Data Index: 274327 (C4 (Web Text))

Max Activating Token Index: 864

Text #12

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2120. Min Act: -0.0000

Data Index: 1408978 (Python Code)

Max Activating Token Index: 729

Text #13

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2071. Min Act: -0.0000

Data Index: 717509 (C4 (Web Text))

Max Activating Token Index: 249

Text #14

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.1995. Min Act: -0.0000

Data Index: 337935 (C4 (Web Text))

Max Activating Token Index: 841

Text #15

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2074. Min Act: -0.0000

Data Index: 502978 (C4 (Web Text))

Max Activating Token Index: 217

Text #16

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2064. Min Act: -0.0000

Data Index: 938596 (C4 (Web Text))

Max Activating Token Index: 515

Text #17

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.2002. Min Act: -0.0000

Data Index: 1307470 (C4 (Web Text))

Max Activating Token Index: 709

Text #18

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.1910. Min Act: -0.0001

Data Index: 720363 (C4 (Web Text))

Max Activating Token Index: 548

Text #19

Max Range: 0.4188. Min Range: -0.4188

Max Act: 0.1962. Min Act: -0.0001

Data Index: 710997 (C4 (Web Text))

Max Activating Token Index: 542