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

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

Neuron 2400 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.2143. Min Act: -0.0001

Data Index: 881977 (C4 (Web Text))

Max Activating Token Index: 469

Text #1

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.2057. Min Act: -0.0001

Data Index: 1528870 (Python Code)

Max Activating Token Index: 682

Text #2

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1911. Min Act: -0.0001

Data Index: 1378603 (Python Code)

Max Activating Token Index: 844

Text #3

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.2012. Min Act: -0.0001

Data Index: 671277 (C4 (Web Text))

Max Activating Token Index: 623

Text #4

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1901. Min Act: -0.0001

Data Index: 700626 (C4 (Web Text))

Max Activating Token Index: 840

Text #5

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.2030. Min Act: -0.0001

Data Index: 1432319 (Python Code)

Max Activating Token Index: 917

Text #6

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1750. Min Act: -0.0001

Data Index: 1552403 (Python Code)

Max Activating Token Index: 438

Text #7

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1844. Min Act: -0.0001

Data Index: 460558 (C4 (Web Text))

Max Activating Token Index: 121

Text #8

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1820. Min Act: -0.0001

Data Index: 1391222 (Python Code)

Max Activating Token Index: 1007

Text #9

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1799. Min Act: -0.0001

Data Index: 234136 (C4 (Web Text))

Max Activating Token Index: 319

Text #10

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1603. Min Act: -0.0001

Data Index: 1440130 (Python Code)

Max Activating Token Index: 631

Text #11

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1842. Min Act: -0.0001

Data Index: 1325311 (C4 (Web Text))

Max Activating Token Index: 330

Text #12

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1751. Min Act: -0.0001

Data Index: 1243629 (C4 (Web Text))

Max Activating Token Index: 177

Text #13

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1723. Min Act: -0.0001

Data Index: 312223 (C4 (Web Text))

Max Activating Token Index: 183

Text #14

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1768. Min Act: -0.0001

Data Index: 1648734 (Python Code)

Max Activating Token Index: 753

Text #15

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1637. Min Act: -0.0001

Data Index: 333200 (C4 (Web Text))

Max Activating Token Index: 1018

Text #16

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1687. Min Act: -0.0001

Data Index: 1536833 (Python Code)

Max Activating Token Index: 400

Text #17

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1661. Min Act: -0.0001

Data Index: 350635 (C4 (Web Text))

Max Activating Token Index: 470

Text #18

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1652. Min Act: -0.0001

Data Index: 715319 (C4 (Web Text))

Max Activating Token Index: 834

Text #19

Max Range: 0.2143. Min Range: -0.2143

Max Act: 0.1644. Min Act: -0.0001

Data Index: 477800 (C4 (Web Text))

Max Activating Token Index: 249