Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3804 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.4324. Min Act: -0.0000

Data Index: 900230 (C4 (Web Text))

Max Activating Token Index: 165

Text #1

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.3792. Min Act: -0.0000

Data Index: 98560 (C4 (Web Text))

Max Activating Token Index: 991

Text #2

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.3754. Min Act: -0.0000

Data Index: 385017 (C4 (Web Text))

Max Activating Token Index: 25

Text #3

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.4139. Min Act: -0.0000

Data Index: 778284 (C4 (Web Text))

Max Activating Token Index: 822

Text #4

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.2876. Min Act: -0.0000

Data Index: 335913 (C4 (Web Text))

Max Activating Token Index: 270

Text #5

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.0228. Min Act: -0.0000

Data Index: 1028490 (C4 (Web Text))

Max Activating Token Index: 468

Text #6

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.1466. Min Act: -0.0000

Data Index: 1115345 (C4 (Web Text))

Max Activating Token Index: 414

Text #7

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.3808. Min Act: -0.0000

Data Index: 912188 (C4 (Web Text))

Max Activating Token Index: 355

Text #8

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.0935. Min Act: -0.0000

Data Index: 789017 (C4 (Web Text))

Max Activating Token Index: 804

Text #9

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.0511. Min Act: -0.0000

Data Index: 1631098 (Python Code)

Max Activating Token Index: 996

Text #10

Max Range: 2.4324. Min Range: -2.4324

Max Act: 2.0563. Min Act: -0.0000

Data Index: 147572 (C4 (Web Text))

Max Activating Token Index: 462

Text #11

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.9972. Min Act: -0.0000

Data Index: 939478 (C4 (Web Text))

Max Activating Token Index: 355

Text #12

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.9109. Min Act: -0.0000

Data Index: 235413 (C4 (Web Text))

Max Activating Token Index: 866

Text #13

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.8798. Min Act: -0.0000

Data Index: 591510 (C4 (Web Text))

Max Activating Token Index: 809

Text #14

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.9067. Min Act: -0.0000

Data Index: 776330 (C4 (Web Text))

Max Activating Token Index: 881

Text #15

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.7253. Min Act: -0.0000

Data Index: 756112 (C4 (Web Text))

Max Activating Token Index: 247

Text #16

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.7251. Min Act: -0.0000

Data Index: 542995 (C4 (Web Text))

Max Activating Token Index: 686

Text #17

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.7585. Min Act: -0.0000

Data Index: 1390017 (Python Code)

Max Activating Token Index: 903

Text #18

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.8116. Min Act: -0.0000

Data Index: 943746 (C4 (Web Text))

Max Activating Token Index: 212

Text #19

Max Range: 2.4324. Min Range: -2.4324

Max Act: 1.6942. Min Act: -0.0000

Data Index: 808070 (C4 (Web Text))

Max Activating Token Index: 259