Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 82 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1056. Min Act: -0.0002

Data Index: 785984 (C4 (Web Text))

Max Activating Token Index: 80

Text #1

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1048. Min Act: -0.0002

Data Index: 521469 (C4 (Web Text))

Max Activating Token Index: 945

Text #2

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1030. Min Act: -0.0002

Data Index: 342276 (C4 (Web Text))

Max Activating Token Index: 46

Text #3

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1079. Min Act: -0.0002

Data Index: 1162236 (C4 (Web Text))

Max Activating Token Index: 606

Text #4

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0989. Min Act: -0.0002

Data Index: 641171 (C4 (Web Text))

Max Activating Token Index: 828

Text #5

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1037. Min Act: -0.0002

Data Index: 268126 (C4 (Web Text))

Max Activating Token Index: 829

Text #6

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1051. Min Act: -0.0002

Data Index: 522624 (C4 (Web Text))

Max Activating Token Index: 828

Text #7

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0994. Min Act: -0.0002

Data Index: 113508 (C4 (Web Text))

Max Activating Token Index: 344

Text #8

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0952. Min Act: -0.0002

Data Index: 912659 (C4 (Web Text))

Max Activating Token Index: 488

Text #9

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0955. Min Act: -0.0002

Data Index: 334937 (C4 (Web Text))

Max Activating Token Index: 292

Text #10

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0994. Min Act: -0.0002

Data Index: 416180 (C4 (Web Text))

Max Activating Token Index: 667

Text #11

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1009. Min Act: -0.0002

Data Index: 1148137 (C4 (Web Text))

Max Activating Token Index: 197

Text #12

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0999. Min Act: -0.0002

Data Index: 22518 (C4 (Web Text))

Max Activating Token Index: 780

Text #13

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1043. Min Act: -0.0002

Data Index: 890235 (C4 (Web Text))

Max Activating Token Index: 265

Text #14

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0974. Min Act: -0.0002

Data Index: 59778 (C4 (Web Text))

Max Activating Token Index: 695

Text #15

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.1005. Min Act: -0.0002

Data Index: 789588 (C4 (Web Text))

Max Activating Token Index: 614

Text #16

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0963. Min Act: -0.0002

Data Index: 206399 (C4 (Web Text))

Max Activating Token Index: 662

Text #17

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0962. Min Act: -0.0002

Data Index: 1334226 (C4 (Web Text))

Max Activating Token Index: 740

Text #18

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0990. Min Act: -0.0002

Data Index: 279496 (C4 (Web Text))

Max Activating Token Index: 594

Text #19

Max Range: 0.1079. Min Range: -0.1079

Max Act: 0.0962. Min Act: -0.0002

Data Index: 344958 (C4 (Web Text))

Max Activating Token Index: 748