Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1567 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.6744. Min Act: -0.0000

Data Index: 428931 (C4 (Web Text))

Max Activating Token Index: 930

Text #1

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.6745. Min Act: -0.0000

Data Index: 457776 (C4 (Web Text))

Max Activating Token Index: 118

Text #2

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.6895. Min Act: -0.0000

Data Index: 253115 (C4 (Web Text))

Max Activating Token Index: 147

Text #3

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.5865. Min Act: -0.0000

Data Index: 378993 (C4 (Web Text))

Max Activating Token Index: 761

Text #4

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.6201. Min Act: -0.0000

Data Index: 1136009 (C4 (Web Text))

Max Activating Token Index: 848

Text #5

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.6455. Min Act: -0.0000

Data Index: 823625 (C4 (Web Text))

Max Activating Token Index: 913

Text #6

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.6215. Min Act: -0.0000

Data Index: 988886 (C4 (Web Text))

Max Activating Token Index: 994

Text #7

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.5713. Min Act: -0.0000

Data Index: 885916 (C4 (Web Text))

Max Activating Token Index: 763

Text #8

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.5854. Min Act: -0.0000

Data Index: 261843 (C4 (Web Text))

Max Activating Token Index: 889

Text #9

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.5612. Min Act: -0.0000

Data Index: 895400 (C4 (Web Text))

Max Activating Token Index: 776

Text #10

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4900. Min Act: -0.0000

Data Index: 300814 (C4 (Web Text))

Max Activating Token Index: 127

Text #11

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.5020. Min Act: -0.0000

Data Index: 1211136 (C4 (Web Text))

Max Activating Token Index: 606

Text #12

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4658. Min Act: -0.0000

Data Index: 876746 (C4 (Web Text))

Max Activating Token Index: 598

Text #13

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4843. Min Act: -0.0000

Data Index: 499176 (C4 (Web Text))

Max Activating Token Index: 245

Text #14

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4700. Min Act: -0.0000

Data Index: 286226 (C4 (Web Text))

Max Activating Token Index: 105

Text #15

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4686. Min Act: -0.0000

Data Index: 536469 (C4 (Web Text))

Max Activating Token Index: 112

Text #16

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4692. Min Act: -0.0000

Data Index: 665854 (C4 (Web Text))

Max Activating Token Index: 596

Text #17

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4527. Min Act: -0.0000

Data Index: 1420693 (Python Code)

Max Activating Token Index: 950

Text #18

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4543. Min Act: -0.0000

Data Index: 772833 (C4 (Web Text))

Max Activating Token Index: 687

Text #19

Max Range: 0.6895. Min Range: -0.6895

Max Act: 0.4462. Min Act: -0.0000

Data Index: 1171254 (C4 (Web Text))

Max Activating Token Index: 516