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

Dataset: The Pile

Neuron 3910 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.8203. Min Act: -0.0000

Data Index: 1439685 (The Pile)

Max Activating Token Index: 437

Text #1

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.8542. Min Act: -0.0000

Data Index: 392307 (The Pile)

Max Activating Token Index: 987

Text #2

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.4427. Min Act: -0.0000

Data Index: 586685 (The Pile)

Max Activating Token Index: 261

Text #3

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.4636. Min Act: -0.0000

Data Index: 1389010 (The Pile)

Max Activating Token Index: 115

Text #4

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.4879. Min Act: -0.0000

Data Index: 79261 (The Pile)

Max Activating Token Index: 110

Text #5

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.0917. Min Act: -0.0000

Data Index: 1170447 (The Pile)

Max Activating Token Index: 910

Text #6

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.2012. Min Act: -0.0000

Data Index: 294827 (The Pile)

Max Activating Token Index: 330

Text #7

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.3621. Min Act: -0.0000

Data Index: 1427033 (The Pile)

Max Activating Token Index: 120

Text #8

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.3392. Min Act: -0.0000

Data Index: 1289487 (The Pile)

Max Activating Token Index: 497

Text #9

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.1044. Min Act: -0.0000

Data Index: 1376265 (The Pile)

Max Activating Token Index: 354

Text #10

Max Range: 3.8542. Min Range: -3.8542

Max Act: 2.9554. Min Act: -0.0000

Data Index: 857495 (The Pile)

Max Activating Token Index: 934

Text #11

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.1802. Min Act: -0.0000

Data Index: 400421 (The Pile)

Max Activating Token Index: 581

Text #12

Max Range: 3.8542. Min Range: -3.8542

Max Act: 2.9583. Min Act: -0.0000

Data Index: 36744 (The Pile)

Max Activating Token Index: 939

Text #13

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.2862. Min Act: -0.0000

Data Index: 496623 (The Pile)

Max Activating Token Index: 162

Text #14

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.1000. Min Act: 0.0000

Data Index: 429286 (The Pile)

Max Activating Token Index: 750

Text #15

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.1547. Min Act: -0.0000

Data Index: 1266138 (The Pile)

Max Activating Token Index: 996

Text #16

Max Range: 3.8542. Min Range: -3.8542

Max Act: 2.7708. Min Act: 0.0000

Data Index: 712347 (The Pile)

Max Activating Token Index: 46

Text #17

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.0284. Min Act: -0.0000

Data Index: 1685445 (The Pile)

Max Activating Token Index: 967

Text #18

Max Range: 3.8542. Min Range: -3.8542

Max Act: 2.8975. Min Act: -0.0000

Data Index: 1956337 (The Pile)

Max Activating Token Index: 847

Text #19

Max Range: 3.8542. Min Range: -3.8542

Max Act: 3.2257. Min Act: -0.0000

Data Index: 988588 (The Pile)

Max Activating Token Index: 870