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

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

Neuron 151 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.6046. Min Act: -0.0001

Data Index: 410010 (C4 (Web Text))

Max Activating Token Index: 135

Text #1

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.2390. Min Act: -0.0001

Data Index: 1596199 (Python Code)

Max Activating Token Index: 53

Text #2

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.3107. Min Act: -0.0001

Data Index: 262453 (C4 (Web Text))

Max Activating Token Index: 833

Text #3

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.3004. Min Act: -0.0001

Data Index: 677352 (C4 (Web Text))

Max Activating Token Index: 19

Text #4

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1885. Min Act: -0.0001

Data Index: 200380 (C4 (Web Text))

Max Activating Token Index: 538

Text #5

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.2140. Min Act: -0.0001

Data Index: 1094063 (C4 (Web Text))

Max Activating Token Index: 285

Text #6

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1688. Min Act: -0.0001

Data Index: 36176 (C4 (Web Text))

Max Activating Token Index: 1016

Text #7

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1588. Min Act: -0.0001

Data Index: 737915 (C4 (Web Text))

Max Activating Token Index: 24

Text #8

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1897. Min Act: -0.0001

Data Index: 402455 (C4 (Web Text))

Max Activating Token Index: 360

Text #9

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1673. Min Act: -0.0001

Data Index: 1476777 (Python Code)

Max Activating Token Index: 235

Text #10

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1176. Min Act: -0.0001

Data Index: 1212128 (C4 (Web Text))

Max Activating Token Index: 729

Text #11

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0355. Min Act: -0.0001

Data Index: 127656 (C4 (Web Text))

Max Activating Token Index: 636

Text #12

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0354. Min Act: -0.0001

Data Index: 1502668 (Python Code)

Max Activating Token Index: 686

Text #13

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0944. Min Act: -0.0001

Data Index: 35763 (C4 (Web Text))

Max Activating Token Index: 540

Text #14

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0587. Min Act: -0.0001

Data Index: 322121 (C4 (Web Text))

Max Activating Token Index: 576

Text #15

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.1134. Min Act: -0.0001

Data Index: 1248971 (C4 (Web Text))

Max Activating Token Index: 883

Text #16

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0812. Min Act: -0.0001

Data Index: 1391797 (Python Code)

Max Activating Token Index: 74

Text #17

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0597. Min Act: -0.0001

Data Index: 218482 (C4 (Web Text))

Max Activating Token Index: 457

Text #18

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0537. Min Act: -0.0001

Data Index: 461196 (C4 (Web Text))

Max Activating Token Index: 19

Text #19

Max Range: 1.6046. Min Range: -1.6046

Max Act: 1.0559. Min Act: -0.0001

Data Index: 287681 (C4 (Web Text))

Max Activating Token Index: 818