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

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

Neuron 3520 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.7995. Min Act: -0.0000

Data Index: 1637513 (Python Code)

Max Activating Token Index: 673

Text #1

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.7077. Min Act: -0.0000

Data Index: 1620790 (Python Code)

Max Activating Token Index: 119

Text #2

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.6801. Min Act: -0.0000

Data Index: 1563165 (Python Code)

Max Activating Token Index: 536

Text #3

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.6870. Min Act: -0.0000

Data Index: 552817 (C4 (Web Text))

Max Activating Token Index: 114

Text #4

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.6264. Min Act: -0.0000

Data Index: 1642577 (Python Code)

Max Activating Token Index: 410

Text #5

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5866. Min Act: -0.0000

Data Index: 1641261 (Python Code)

Max Activating Token Index: 577

Text #6

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.6188. Min Act: -0.0000

Data Index: 331985 (C4 (Web Text))

Max Activating Token Index: 315

Text #7

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5755. Min Act: -0.0000

Data Index: 1465115 (Python Code)

Max Activating Token Index: 545

Text #8

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.6087. Min Act: -0.0000

Data Index: 250301 (C4 (Web Text))

Max Activating Token Index: 1002

Text #9

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5891. Min Act: -0.0000

Data Index: 1407633 (Python Code)

Max Activating Token Index: 601

Text #10

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5598. Min Act: -0.0000

Data Index: 1393629 (Python Code)

Max Activating Token Index: 1001

Text #11

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5961. Min Act: -0.0000

Data Index: 1385054 (Python Code)

Max Activating Token Index: 185

Text #12

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5794. Min Act: -0.0000

Data Index: 1376305 (Python Code)

Max Activating Token Index: 311

Text #13

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5574. Min Act: -0.0000

Data Index: 1469122 (Python Code)

Max Activating Token Index: 72

Text #14

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5641. Min Act: -0.0000

Data Index: 1592431 (Python Code)

Max Activating Token Index: 370

Text #15

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5248. Min Act: -0.0000

Data Index: 77112 (C4 (Web Text))

Max Activating Token Index: 241

Text #16

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5359. Min Act: -0.0000

Data Index: 899181 (C4 (Web Text))

Max Activating Token Index: 849

Text #17

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5311. Min Act: -0.0000

Data Index: 1575366 (Python Code)

Max Activating Token Index: 222

Text #18

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5106. Min Act: -0.0000

Data Index: 1549501 (Python Code)

Max Activating Token Index: 89

Text #19

Max Range: 0.7995. Min Range: -0.7995

Max Act: 0.5152. Min Act: -0.0000

Data Index: 1524853 (Python Code)

Max Activating Token Index: 157