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

Dataset: The Pile

Neuron 1382 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.3921. Min Act: -0.0000

Data Index: 1265958 (The Pile)

Max Activating Token Index: 760

Text #1

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.4927. Min Act: -0.0000

Data Index: 121539 (The Pile)

Max Activating Token Index: 668

Text #2

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.2962. Min Act: -0.0000

Data Index: 1232494 (The Pile)

Max Activating Token Index: 398

Text #3

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.6273. Min Act: -0.0000

Data Index: 1661690 (The Pile)

Max Activating Token Index: 1005

Text #4

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.1802. Min Act: -0.0000

Data Index: 1891804 (The Pile)

Max Activating Token Index: 767

Text #5

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.9567. Min Act: -0.0000

Data Index: 1536635 (The Pile)

Max Activating Token Index: 871

Text #6

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.9584. Min Act: -0.0000

Data Index: 1171559 (The Pile)

Max Activating Token Index: 907

Text #7

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.7418. Min Act: -0.0000

Data Index: 1754238 (The Pile)

Max Activating Token Index: 770

Text #8

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.0256. Min Act: -0.0000

Data Index: 427833 (The Pile)

Max Activating Token Index: 709

Text #9

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.0519. Min Act: -0.0000

Data Index: 1924725 (The Pile)

Max Activating Token Index: 837

Text #10

Max Range: 8.4927. Min Range: -8.4927

Max Act: 8.1135. Min Act: -0.0000

Data Index: 1650386 (The Pile)

Max Activating Token Index: 704

Text #11

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.7077. Min Act: -0.0000

Data Index: 1742732 (The Pile)

Max Activating Token Index: 783

Text #12

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.9430. Min Act: -0.0000

Data Index: 1639744 (The Pile)

Max Activating Token Index: 204

Text #13

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.8741. Min Act: -0.0000

Data Index: 1729217 (The Pile)

Max Activating Token Index: 531

Text #14

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.6015. Min Act: -0.0000

Data Index: 987247 (The Pile)

Max Activating Token Index: 490

Text #15

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.3599. Min Act: -0.0000

Data Index: 1662066 (The Pile)

Max Activating Token Index: 448

Text #16

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.7198. Min Act: -0.0000

Data Index: 1255632 (The Pile)

Max Activating Token Index: 326

Text #17

Max Range: 8.4927. Min Range: -8.4927

Max Act: 6.9236. Min Act: -0.0000

Data Index: 68911 (The Pile)

Max Activating Token Index: 525

Text #18

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.4735. Min Act: -0.0000

Data Index: 1203690 (The Pile)

Max Activating Token Index: 776

Text #19

Max Range: 8.4927. Min Range: -8.4927

Max Act: 7.0123. Min Act: -0.0000

Data Index: 259522 (The Pile)

Max Activating Token Index: 484