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

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

Neuron 3542 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.7509. Min Act: -0.0000

Data Index: 1469817 (Python Code)

Max Activating Token Index: 70

Text #1

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.7071. Min Act: -0.0000

Data Index: 1624936 (Python Code)

Max Activating Token Index: 145

Text #2

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.7147. Min Act: -0.0000

Data Index: 1553809 (Python Code)

Max Activating Token Index: 134

Text #3

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6447. Min Act: -0.0000

Data Index: 186749 (C4 (Web Text))

Max Activating Token Index: 852

Text #4

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6667. Min Act: -0.0000

Data Index: 1499834 (Python Code)

Max Activating Token Index: 144

Text #5

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6095. Min Act: -0.0000

Data Index: 1408125 (Python Code)

Max Activating Token Index: 329

Text #6

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6663. Min Act: -0.0000

Data Index: 1055624 (C4 (Web Text))

Max Activating Token Index: 290

Text #7

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6598. Min Act: -0.0000

Data Index: 1230456 (C4 (Web Text))

Max Activating Token Index: 72

Text #8

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6136. Min Act: -0.0000

Data Index: 1370358 (Python Code)

Max Activating Token Index: 217

Text #9

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6245. Min Act: -0.0000

Data Index: 1431391 (Python Code)

Max Activating Token Index: 49

Text #10

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6221. Min Act: -0.0000

Data Index: 1569138 (Python Code)

Max Activating Token Index: 3

Text #11

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6010. Min Act: -0.0000

Data Index: 1397580 (Python Code)

Max Activating Token Index: 99

Text #12

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5756. Min Act: -0.0001

Data Index: 999320 (C4 (Web Text))

Max Activating Token Index: 562

Text #13

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.6072. Min Act: -0.0000

Data Index: 619619 (C4 (Web Text))

Max Activating Token Index: 592

Text #14

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5498. Min Act: -0.0000

Data Index: 1643707 (Python Code)

Max Activating Token Index: 195

Text #15

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5850. Min Act: -0.0000

Data Index: 1506834 (Python Code)

Max Activating Token Index: 18

Text #16

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5842. Min Act: -0.0000

Data Index: 1546211 (Python Code)

Max Activating Token Index: 790

Text #17

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5767. Min Act: -0.0000

Data Index: 1437558 (Python Code)

Max Activating Token Index: 23

Text #18

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5923. Min Act: -0.0000

Data Index: 1428745 (Python Code)

Max Activating Token Index: 128

Text #19

Max Range: 0.7509. Min Range: -0.7509

Max Act: 0.5878. Min Act: -0.0001

Data Index: 253123 (C4 (Web Text))

Max Activating Token Index: 62