Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4622 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.3065. Min Range: -2.3065

Max Act: 2.3065. Min Act: -0.0000

Data Index: 794243 (C4 (Web Text))

Max Activating Token Index: 813

Text #1

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.9717. Min Act: -0.0000

Data Index: 1139486 (C4 (Web Text))

Max Activating Token Index: 676

Text #2

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.9891. Min Act: -0.0000

Data Index: 1026177 (C4 (Web Text))

Max Activating Token Index: 385

Text #3

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.8403. Min Act: -0.0000

Data Index: 157869 (C4 (Web Text))

Max Activating Token Index: 986

Text #4

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.8179. Min Act: -0.0000

Data Index: 1258285 (C4 (Web Text))

Max Activating Token Index: 920

Text #5

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.8098. Min Act: -0.0000

Data Index: 781946 (C4 (Web Text))

Max Activating Token Index: 422

Text #6

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.7196. Min Act: -0.0000

Data Index: 719908 (C4 (Web Text))

Max Activating Token Index: 918

Text #7

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.7592. Min Act: -0.0000

Data Index: 1141767 (C4 (Web Text))

Max Activating Token Index: 1001

Text #8

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.8129. Min Act: -0.0000

Data Index: 1215443 (C4 (Web Text))

Max Activating Token Index: 246

Text #9

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.6955. Min Act: -0.0000

Data Index: 881825 (C4 (Web Text))

Max Activating Token Index: 647

Text #10

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.7035. Min Act: -0.0000

Data Index: 55459 (C4 (Web Text))

Max Activating Token Index: 928

Text #11

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.6954. Min Act: -0.0000

Data Index: 822360 (C4 (Web Text))

Max Activating Token Index: 439

Text #12

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.6775. Min Act: -0.0000

Data Index: 810162 (C4 (Web Text))

Max Activating Token Index: 523

Text #13

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.5807. Min Act: -0.0000

Data Index: 1147024 (C4 (Web Text))

Max Activating Token Index: 866

Text #14

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.5538. Min Act: -0.0000

Data Index: 1245562 (C4 (Web Text))

Max Activating Token Index: 137

Text #15

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.6443. Min Act: -0.0000

Data Index: 633219 (C4 (Web Text))

Max Activating Token Index: 523

Text #16

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.5977. Min Act: -0.0000

Data Index: 1461989 (Python Code)

Max Activating Token Index: 765

Text #17

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.5908. Min Act: -0.0000

Data Index: 1290213 (C4 (Web Text))

Max Activating Token Index: 644

Text #18

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.5694. Min Act: -0.0000

Data Index: 123368 (C4 (Web Text))

Max Activating Token Index: 101

Text #19

Max Range: 2.3065. Min Range: -2.3065

Max Act: 1.4813. Min Act: -0.0000

Data Index: 859426 (C4 (Web Text))

Max Activating Token Index: 432