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

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

Neuron 1650 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: 0.1015. Min Range: -0.1015

Max Act: 0.1007. Min Act: -0.0001

Data Index: 813698 (C4 (Web Text))

Max Activating Token Index: 467

Text #1

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0930. Min Act: -0.0001

Data Index: 551440 (C4 (Web Text))

Max Activating Token Index: 290

Text #2

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0836. Min Act: -0.0001

Data Index: 89386 (C4 (Web Text))

Max Activating Token Index: 302

Text #3

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0950. Min Act: -0.0001

Data Index: 98627 (C4 (Web Text))

Max Activating Token Index: 260

Text #4

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.1015. Min Act: -0.0001

Data Index: 71509 (C4 (Web Text))

Max Activating Token Index: 629

Text #5

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0850. Min Act: -0.0001

Data Index: 853997 (C4 (Web Text))

Max Activating Token Index: 165

Text #6

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0829. Min Act: -0.0001

Data Index: 1206080 (C4 (Web Text))

Max Activating Token Index: 295

Text #7

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0828. Min Act: -0.0001

Data Index: 1042727 (C4 (Web Text))

Max Activating Token Index: 835

Text #8

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0763. Min Act: -0.0001

Data Index: 521725 (C4 (Web Text))

Max Activating Token Index: 863

Text #9

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0740. Min Act: -0.0001

Data Index: 742386 (C4 (Web Text))

Max Activating Token Index: 593

Text #10

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0833. Min Act: -0.0001

Data Index: 1109198 (C4 (Web Text))

Max Activating Token Index: 163

Text #11

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0813. Min Act: -0.0001

Data Index: 1013643 (C4 (Web Text))

Max Activating Token Index: 505

Text #12

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0814. Min Act: -0.0001

Data Index: 275448 (C4 (Web Text))

Max Activating Token Index: 888

Text #13

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0746. Min Act: -0.0001

Data Index: 509270 (C4 (Web Text))

Max Activating Token Index: 601

Text #14

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0748. Min Act: -0.0001

Data Index: 1242084 (C4 (Web Text))

Max Activating Token Index: 365

Text #15

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0736. Min Act: -0.0001

Data Index: 108589 (C4 (Web Text))

Max Activating Token Index: 697

Text #16

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0859. Min Act: -0.0001

Data Index: 687500 (C4 (Web Text))

Max Activating Token Index: 232

Text #17

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0877. Min Act: -0.0001

Data Index: 213233 (C4 (Web Text))

Max Activating Token Index: 385

Text #18

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0711. Min Act: -0.0001

Data Index: 1311920 (C4 (Web Text))

Max Activating Token Index: 240

Text #19

Max Range: 0.1015. Min Range: -0.1015

Max Act: 0.0723. Min Act: -0.0001

Data Index: 1431334 (Python Code)

Max Activating Token Index: 84