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

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

Neuron 1559 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6955. Min Act: -0.0001

Data Index: 430458 (C4 (Web Text))

Max Activating Token Index: 540

Text #1

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6496. Min Act: -0.0000

Data Index: 182168 (C4 (Web Text))

Max Activating Token Index: 148

Text #2

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6870. Min Act: -0.0000

Data Index: 578810 (C4 (Web Text))

Max Activating Token Index: 81

Text #3

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6623. Min Act: -0.0000

Data Index: 535942 (C4 (Web Text))

Max Activating Token Index: 338

Text #4

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6573. Min Act: -0.0000

Data Index: 233440 (C4 (Web Text))

Max Activating Token Index: 190

Text #5

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6475. Min Act: -0.0000

Data Index: 981351 (C4 (Web Text))

Max Activating Token Index: 313

Text #6

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6210. Min Act: -0.0001

Data Index: 68147 (C4 (Web Text))

Max Activating Token Index: 158

Text #7

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6181. Min Act: -0.0001

Data Index: 353905 (C4 (Web Text))

Max Activating Token Index: 316

Text #8

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5947. Min Act: -0.0000

Data Index: 332642 (C4 (Web Text))

Max Activating Token Index: 140

Text #9

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5839. Min Act: -0.0000

Data Index: 738860 (C4 (Web Text))

Max Activating Token Index: 911

Text #10

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6230. Min Act: -0.0000

Data Index: 464797 (C4 (Web Text))

Max Activating Token Index: 420

Text #11

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6002. Min Act: -0.0000

Data Index: 477725 (C4 (Web Text))

Max Activating Token Index: 531

Text #12

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5932. Min Act: -0.0000

Data Index: 687166 (C4 (Web Text))

Max Activating Token Index: 788

Text #13

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.6135. Min Act: -0.0000

Data Index: 161666 (C4 (Web Text))

Max Activating Token Index: 826

Text #14

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5869. Min Act: -0.0000

Data Index: 809299 (C4 (Web Text))

Max Activating Token Index: 67

Text #15

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5984. Min Act: -0.0000

Data Index: 967205 (C4 (Web Text))

Max Activating Token Index: 438

Text #16

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5406. Min Act: -0.0001

Data Index: 1569955 (Python Code)

Max Activating Token Index: 891

Text #17

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5736. Min Act: -0.0000

Data Index: 817015 (C4 (Web Text))

Max Activating Token Index: 418

Text #18

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5929. Min Act: -0.0000

Data Index: 767576 (C4 (Web Text))

Max Activating Token Index: 287

Text #19

Max Range: 0.6955. Min Range: -0.6955

Max Act: 0.5710. Min Act: -0.0000

Data Index: 420789 (C4 (Web Text))

Max Activating Token Index: 729