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

Dataset: The Pile

Neuron 304 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.9235. Min Act: -0.0000

Data Index: 1111189 (The Pile)

Max Activating Token Index: 532

Text #1

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.8941. Min Act: -0.0000

Data Index: 267544 (The Pile)

Max Activating Token Index: 523

Text #2

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.3751. Min Act: -0.0000

Data Index: 1976605 (The Pile)

Max Activating Token Index: 224

Text #3

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.5919. Min Act: -0.0000

Data Index: 943834 (The Pile)

Max Activating Token Index: 486

Text #4

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.1861. Min Act: -0.0000

Data Index: 1836429 (The Pile)

Max Activating Token Index: 644

Text #5

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.5529. Min Act: -0.0000

Data Index: 587051 (The Pile)

Max Activating Token Index: 484

Text #6

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.3294. Min Act: -0.0000

Data Index: 1166322 (The Pile)

Max Activating Token Index: 144

Text #7

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0944. Min Act: -0.0000

Data Index: 991977 (The Pile)

Max Activating Token Index: 737

Text #8

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.6501. Min Act: -0.0000

Data Index: 751044 (The Pile)

Max Activating Token Index: 435

Text #9

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0699. Min Act: -0.0000

Data Index: 431934 (The Pile)

Max Activating Token Index: 694

Text #10

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.1196. Min Act: -0.0000

Data Index: 1272757 (The Pile)

Max Activating Token Index: 340

Text #11

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.1954. Min Act: -0.0000

Data Index: 1507207 (The Pile)

Max Activating Token Index: 795

Text #12

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.2099. Min Act: -0.0000

Data Index: 602448 (The Pile)

Max Activating Token Index: 720

Text #13

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0984. Min Act: -0.0000

Data Index: 1551092 (The Pile)

Max Activating Token Index: 742

Text #14

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0500. Min Act: -0.0000

Data Index: 1157156 (The Pile)

Max Activating Token Index: 906

Text #15

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0633. Min Act: -0.0000

Data Index: 1308388 (The Pile)

Max Activating Token Index: 440

Text #16

Max Range: 3.9235. Min Range: -3.9235

Max Act: 2.9072. Min Act: -0.0000

Data Index: 1589567 (The Pile)

Max Activating Token Index: 506

Text #17

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0537. Min Act: -0.0000

Data Index: 1292083 (The Pile)

Max Activating Token Index: 447

Text #18

Max Range: 3.9235. Min Range: -3.9235

Max Act: 2.9865. Min Act: -0.0000

Data Index: 626497 (The Pile)

Max Activating Token Index: 160

Text #19

Max Range: 3.9235. Min Range: -3.9235

Max Act: 3.0337. Min Act: -0.0000

Data Index: 232221 (The Pile)

Max Activating Token Index: 415