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

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

Neuron 788 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0946. Min Act: -0.0002

Data Index: 226393 (C4 (Web Text))

Max Activating Token Index: 171

Text #1

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0897. Min Act: -0.0001

Data Index: 229127 (C4 (Web Text))

Max Activating Token Index: 84

Text #2

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0892. Min Act: -0.0001

Data Index: 940864 (C4 (Web Text))

Max Activating Token Index: 4

Text #3

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0915. Min Act: -0.0001

Data Index: 279699 (C4 (Web Text))

Max Activating Token Index: 84

Text #4

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0820. Min Act: -0.0001

Data Index: 543734 (C4 (Web Text))

Max Activating Token Index: 909

Text #5

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0928. Min Act: -0.0001

Data Index: 821049 (C4 (Web Text))

Max Activating Token Index: 881

Text #6

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0873. Min Act: -0.0002

Data Index: 266122 (C4 (Web Text))

Max Activating Token Index: 216

Text #7

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0801. Min Act: -0.0001

Data Index: 155754 (C4 (Web Text))

Max Activating Token Index: 938

Text #8

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0823. Min Act: -0.0001

Data Index: 508103 (C4 (Web Text))

Max Activating Token Index: 801

Text #9

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0820. Min Act: -0.0001

Data Index: 476889 (C4 (Web Text))

Max Activating Token Index: 71

Text #10

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0828. Min Act: -0.0001

Data Index: 289187 (C4 (Web Text))

Max Activating Token Index: 11

Text #11

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0811. Min Act: -0.0001

Data Index: 344138 (C4 (Web Text))

Max Activating Token Index: 612

Text #12

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0821. Min Act: -0.0002

Data Index: 1108861 (C4 (Web Text))

Max Activating Token Index: 992

Text #13

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0846. Min Act: -0.0002

Data Index: 607296 (C4 (Web Text))

Max Activating Token Index: 48

Text #14

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0777. Min Act: -0.0002

Data Index: 271854 (C4 (Web Text))

Max Activating Token Index: 973

Text #15

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0696. Min Act: -0.0001

Data Index: 1549717 (Python Code)

Max Activating Token Index: 580

Text #16

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0787. Min Act: -0.0002

Data Index: 8154 (C4 (Web Text))

Max Activating Token Index: 501

Text #17

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0832. Min Act: -0.0001

Data Index: 422984 (C4 (Web Text))

Max Activating Token Index: 1009

Text #18

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0750. Min Act: -0.0001

Data Index: 999085 (C4 (Web Text))

Max Activating Token Index: 26

Text #19

Max Range: 0.0946. Min Range: -0.0946

Max Act: 0.0744. Min Act: -0.0002

Data Index: 233322 (C4 (Web Text))

Max Activating Token Index: 895