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

Dataset: The Pile

Neuron 1469 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.2898. Min Act: -0.0000

Data Index: 1561570 (The Pile)

Max Activating Token Index: 664

Text #1

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.3526. Min Act: -0.0000

Data Index: 1944814 (The Pile)

Max Activating Token Index: 751

Text #2

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.3591. Min Act: -0.0000

Data Index: 1672299 (The Pile)

Max Activating Token Index: 937

Text #3

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.4227. Min Act: -0.0000

Data Index: 592236 (The Pile)

Max Activating Token Index: 593

Text #4

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.2066. Min Act: -0.0000

Data Index: 1372512 (The Pile)

Max Activating Token Index: 845

Text #5

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.0519. Min Act: -0.0000

Data Index: 1194914 (The Pile)

Max Activating Token Index: 654

Text #6

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.0910. Min Act: -0.0000

Data Index: 1327823 (The Pile)

Max Activating Token Index: 825

Text #7

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.1541. Min Act: -0.0000

Data Index: 1342213 (The Pile)

Max Activating Token Index: 863

Text #8

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.2135. Min Act: -0.0000

Data Index: 775875 (The Pile)

Max Activating Token Index: 591

Text #9

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.2127. Min Act: -0.0000

Data Index: 1205359 (The Pile)

Max Activating Token Index: 625

Text #10

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.2801. Min Act: -0.0000

Data Index: 1883407 (The Pile)

Max Activating Token Index: 861

Text #11

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.0695. Min Act: -0.0000

Data Index: 159173 (The Pile)

Max Activating Token Index: 812

Text #12

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.1389. Min Act: -0.0000

Data Index: 292176 (The Pile)

Max Activating Token Index: 439

Text #13

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.1616. Min Act: -0.0000

Data Index: 869840 (The Pile)

Max Activating Token Index: 130

Text #14

Max Range: 13.4227. Min Range: -13.4227

Max Act: 12.8527. Min Act: -0.0000

Data Index: 880332 (The Pile)

Max Activating Token Index: 735

Text #15

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.2174. Min Act: -0.0000

Data Index: 875847 (The Pile)

Max Activating Token Index: 706

Text #16

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.0661. Min Act: -0.0000

Data Index: 1389746 (The Pile)

Max Activating Token Index: 738

Text #17

Max Range: 13.4227. Min Range: -13.4227

Max Act: 13.0009. Min Act: -0.0000

Data Index: 512605 (The Pile)

Max Activating Token Index: 875

Text #18

Max Range: 13.4227. Min Range: -13.4227

Max Act: 12.9933. Min Act: -0.0000

Data Index: 343628 (The Pile)

Max Activating Token Index: 661

Text #19

Max Range: 13.4227. Min Range: -13.4227

Max Act: 12.8444. Min Act: -0.0000

Data Index: 1688139 (The Pile)

Max Activating Token Index: 806