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

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

Neuron 137 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.1110. Min Act: -0.0005

Data Index: 1612982 (Python Code)

Max Activating Token Index: 952

Text #1

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0986. Min Act: -0.0005

Data Index: 198293 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0894. Min Act: -0.0005

Data Index: 89487 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0761. Min Act: -0.0005

Data Index: 999809 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0908. Min Act: -0.0005

Data Index: 195886 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0943. Min Act: -0.0005

Data Index: 1190917 (C4 (Web Text))

Max Activating Token Index: 869

Text #6

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0923. Min Act: -0.0005

Data Index: 146111 (C4 (Web Text))

Max Activating Token Index: 1000

Text #7

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0892. Min Act: -0.0005

Data Index: 955089 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0890. Min Act: -0.0005

Data Index: 941706 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0756. Min Act: -0.0005

Data Index: 900198 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0896. Min Act: -0.0005

Data Index: 1070840 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0910. Min Act: -0.0005

Data Index: 60248 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0863. Min Act: -0.0005

Data Index: 802118 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0837. Min Act: -0.0005

Data Index: 9417 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0828. Min Act: -0.0005

Data Index: 851666 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0351. Min Act: -0.0005

Data Index: 478991 (C4 (Web Text))

Max Activating Token Index: 1009

Text #16

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0796. Min Act: -0.0005

Data Index: 105675 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0799. Min Act: -0.0005

Data Index: 150478 (C4 (Web Text))

Max Activating Token Index: 629

Text #18

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0815. Min Act: -0.0005

Data Index: 387260 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.1110. Min Range: -0.1110

Max Act: 0.0802. Min Act: -0.0005

Data Index: 701153 (C4 (Web Text))

Max Activating Token Index: 1023