Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3256 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3748. Min Act: -0.0000

Data Index: 72845 (C4 (Web Text))

Max Activating Token Index: 51

Text #1

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3616. Min Act: -0.0001

Data Index: 1188708 (C4 (Web Text))

Max Activating Token Index: 545

Text #2

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3365. Min Act: -0.0000

Data Index: 1094836 (C4 (Web Text))

Max Activating Token Index: 731

Text #3

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3293. Min Act: -0.0000

Data Index: 190142 (C4 (Web Text))

Max Activating Token Index: 194

Text #4

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3077. Min Act: -0.0001

Data Index: 1602064 (Python Code)

Max Activating Token Index: 889

Text #5

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3326. Min Act: -0.0001

Data Index: 1595862 (Python Code)

Max Activating Token Index: 985

Text #6

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3228. Min Act: -0.0001

Data Index: 455425 (C4 (Web Text))

Max Activating Token Index: 655

Text #7

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3013. Min Act: -0.0001

Data Index: 436504 (C4 (Web Text))

Max Activating Token Index: 119

Text #8

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3031. Min Act: -0.0000

Data Index: 131888 (C4 (Web Text))

Max Activating Token Index: 146

Text #9

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2888. Min Act: -0.0001

Data Index: 1543541 (Python Code)

Max Activating Token Index: 998

Text #10

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2778. Min Act: -0.0000

Data Index: 258524 (C4 (Web Text))

Max Activating Token Index: 837

Text #11

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2969. Min Act: -0.0001

Data Index: 1423405 (Python Code)

Max Activating Token Index: 983

Text #12

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2840. Min Act: -0.0000

Data Index: 504264 (C4 (Web Text))

Max Activating Token Index: 657

Text #13

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2896. Min Act: -0.0001

Data Index: 1041920 (C4 (Web Text))

Max Activating Token Index: 686

Text #14

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2974. Min Act: -0.0000

Data Index: 753002 (C4 (Web Text))

Max Activating Token Index: 285

Text #15

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2859. Min Act: -0.0001

Data Index: 1284269 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2766. Min Act: -0.0001

Data Index: 1032159 (C4 (Web Text))

Max Activating Token Index: 794

Text #17

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2846. Min Act: -0.0001

Data Index: 1048704 (C4 (Web Text))

Max Activating Token Index: 285

Text #18

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.2806. Min Act: -0.0000

Data Index: 1061588 (C4 (Web Text))

Max Activating Token Index: 75

Text #19

Max Range: 0.3748. Min Range: -0.3748

Max Act: 0.3007. Min Act: -0.0001

Data Index: 1134716 (C4 (Web Text))

Max Activating Token Index: 666