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

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

Neuron 3102 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0984. Min Act: -0.0001

Data Index: 1621533 (Python Code)

Max Activating Token Index: 119

Text #1

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0925. Min Act: -0.0001

Data Index: 781262 (C4 (Web Text))

Max Activating Token Index: 582

Text #2

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0763. Min Act: -0.0001

Data Index: 648764 (C4 (Web Text))

Max Activating Token Index: 350

Text #3

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0797. Min Act: -0.0001

Data Index: 28835 (C4 (Web Text))

Max Activating Token Index: 9

Text #4

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0802. Min Act: -0.0001

Data Index: 1591474 (Python Code)

Max Activating Token Index: 64

Text #5

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0821. Min Act: -0.0001

Data Index: 289659 (C4 (Web Text))

Max Activating Token Index: 829

Text #6

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0786. Min Act: -0.0001

Data Index: 1147270 (C4 (Web Text))

Max Activating Token Index: 877

Text #7

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0770. Min Act: -0.0001

Data Index: 972159 (C4 (Web Text))

Max Activating Token Index: 494

Text #8

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0783. Min Act: -0.0001

Data Index: 314224 (C4 (Web Text))

Max Activating Token Index: 578

Text #9

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0831. Min Act: -0.0001

Data Index: 946354 (C4 (Web Text))

Max Activating Token Index: 338

Text #10

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0768. Min Act: -0.0001

Data Index: 1299720 (C4 (Web Text))

Max Activating Token Index: 1016

Text #11

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0741. Min Act: -0.0001

Data Index: 847898 (C4 (Web Text))

Max Activating Token Index: 741

Text #12

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0741. Min Act: -0.0001

Data Index: 471195 (C4 (Web Text))

Max Activating Token Index: 660

Text #13

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0718. Min Act: -0.0001

Data Index: 197921 (C4 (Web Text))

Max Activating Token Index: 920

Text #14

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0741. Min Act: -0.0001

Data Index: 1219534 (C4 (Web Text))

Max Activating Token Index: 48

Text #15

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0737. Min Act: -0.0001

Data Index: 1443479 (Python Code)

Max Activating Token Index: 161

Text #16

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0741. Min Act: -0.0001

Data Index: 1279141 (C4 (Web Text))

Max Activating Token Index: 307

Text #17

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0734. Min Act: -0.0001

Data Index: 1318893 (C4 (Web Text))

Max Activating Token Index: 603

Text #18

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0731. Min Act: -0.0001

Data Index: 902431 (C4 (Web Text))

Max Activating Token Index: 942

Text #19

Max Range: 0.0984. Min Range: -0.0984

Max Act: 0.0727. Min Act: -0.0001

Data Index: 738021 (C4 (Web Text))

Max Activating Token Index: 361