Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 25 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0074. Min Act: -0.0006

Data Index: 970304 (C4 (Web Text))

Max Activating Token Index: 29

Text #1

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0072. Min Act: -0.0006

Data Index: 769757 (C4 (Web Text))

Max Activating Token Index: 915

Text #2

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0070. Min Act: -0.0006

Data Index: 1185750 (C4 (Web Text))

Max Activating Token Index: 86

Text #3

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0069. Min Act: -0.0006

Data Index: 528979 (C4 (Web Text))

Max Activating Token Index: 121

Text #4

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0069. Min Act: -0.0006

Data Index: 550972 (C4 (Web Text))

Max Activating Token Index: 1022

Text #5

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0065. Min Act: -0.0005

Data Index: 1630651 (Python Code)

Max Activating Token Index: 701

Text #6

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0067. Min Act: -0.0006

Data Index: 1583046 (Python Code)

Max Activating Token Index: 991

Text #7

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0066. Min Act: -0.0006

Data Index: 250647 (C4 (Web Text))

Max Activating Token Index: 119

Text #8

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0065. Min Act: -0.0006

Data Index: 781523 (C4 (Web Text))

Max Activating Token Index: 749

Text #9

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0066. Min Act: -0.0005

Data Index: 659615 (C4 (Web Text))

Max Activating Token Index: 754

Text #10

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0066. Min Act: -0.0005

Data Index: 353717 (C4 (Web Text))

Max Activating Token Index: 164

Text #11

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0065. Min Act: -0.0006

Data Index: 707986 (C4 (Web Text))

Max Activating Token Index: 21

Text #12

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0064. Min Act: -0.0006

Data Index: 351379 (C4 (Web Text))

Max Activating Token Index: 908

Text #13

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0063. Min Act: -0.0005

Data Index: 1562902 (Python Code)

Max Activating Token Index: 619

Text #14

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0063. Min Act: -0.0006

Data Index: 1495214 (Python Code)

Max Activating Token Index: 148

Text #15

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0063. Min Act: -0.0005

Data Index: 1655582 (Python Code)

Max Activating Token Index: 292

Text #16

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0064. Min Act: -0.0006

Data Index: 625644 (C4 (Web Text))

Max Activating Token Index: 242

Text #17

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0063. Min Act: -0.0006

Data Index: 392433 (C4 (Web Text))

Max Activating Token Index: 436

Text #18

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0062. Min Act: -0.0005

Data Index: 1495836 (Python Code)

Max Activating Token Index: 627

Text #19

Max Range: 0.0074. Min Range: -0.0074

Max Act: 0.0063. Min Act: -0.0006

Data Index: 430029 (C4 (Web Text))

Max Activating Token Index: 79