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

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

Neuron 617 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.2440. Min Act: -0.0003

Data Index: 61796 (C4 (Web Text))

Max Activating Token Index: 792

Text #1

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.2410. Min Act: -0.0003

Data Index: 1059593 (C4 (Web Text))

Max Activating Token Index: 852

Text #2

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1998. Min Act: -0.0004

Data Index: 488852 (C4 (Web Text))

Max Activating Token Index: 757

Text #3

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.2018. Min Act: -0.0003

Data Index: 58243 (C4 (Web Text))

Max Activating Token Index: 629

Text #4

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1863. Min Act: -0.0003

Data Index: 46205 (C4 (Web Text))

Max Activating Token Index: 198

Text #5

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1921. Min Act: -0.0004

Data Index: 531104 (C4 (Web Text))

Max Activating Token Index: 407

Text #6

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1829. Min Act: -0.0003

Data Index: 690531 (C4 (Web Text))

Max Activating Token Index: 691

Text #7

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1736. Min Act: -0.0003

Data Index: 812623 (C4 (Web Text))

Max Activating Token Index: 110

Text #8

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1774. Min Act: -0.0003

Data Index: 702136 (C4 (Web Text))

Max Activating Token Index: 375

Text #9

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1815. Min Act: -0.0003

Data Index: 989917 (C4 (Web Text))

Max Activating Token Index: 792

Text #10

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1784. Min Act: -0.0003

Data Index: 952068 (C4 (Web Text))

Max Activating Token Index: 310

Text #11

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1679. Min Act: -0.0003

Data Index: 1201942 (C4 (Web Text))

Max Activating Token Index: 181

Text #12

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1746. Min Act: -0.0003

Data Index: 174314 (C4 (Web Text))

Max Activating Token Index: 999

Text #13

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1686. Min Act: -0.0003

Data Index: 1131435 (C4 (Web Text))

Max Activating Token Index: 217

Text #14

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1718. Min Act: -0.0003

Data Index: 666125 (C4 (Web Text))

Max Activating Token Index: 20

Text #15

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1654. Min Act: -0.0004

Data Index: 95353 (C4 (Web Text))

Max Activating Token Index: 582

Text #16

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1690. Min Act: -0.0003

Data Index: 821141 (C4 (Web Text))

Max Activating Token Index: 221

Text #17

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1730. Min Act: -0.0003

Data Index: 1090688 (C4 (Web Text))

Max Activating Token Index: 659

Text #18

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1556. Min Act: -0.0004

Data Index: 906093 (C4 (Web Text))

Max Activating Token Index: 496

Text #19

Max Range: 0.2440. Min Range: -0.2440

Max Act: 0.1615. Min Act: -0.0003

Data Index: 401268 (C4 (Web Text))

Max Activating Token Index: 616