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

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

Neuron 3840 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.4553. Min Act: -0.0000

Data Index: 1352012 (C4 (Web Text))

Max Activating Token Index: 336

Text #1

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.4956. Min Act: -0.0000

Data Index: 371976 (C4 (Web Text))

Max Activating Token Index: 742

Text #2

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.5110. Min Act: -0.0000

Data Index: 424488 (C4 (Web Text))

Max Activating Token Index: 144

Text #3

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.4177. Min Act: -0.0000

Data Index: 386030 (C4 (Web Text))

Max Activating Token Index: 500

Text #4

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.4043. Min Act: -0.0000

Data Index: 1267635 (C4 (Web Text))

Max Activating Token Index: 397

Text #5

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3906. Min Act: -0.0000

Data Index: 1368946 (Python Code)

Max Activating Token Index: 948

Text #6

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3981. Min Act: -0.0000

Data Index: 144116 (C4 (Web Text))

Max Activating Token Index: 513

Text #7

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3811. Min Act: -0.0000

Data Index: 492526 (C4 (Web Text))

Max Activating Token Index: 388

Text #8

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3551. Min Act: -0.0000

Data Index: 855014 (C4 (Web Text))

Max Activating Token Index: 746

Text #9

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3668. Min Act: -0.0000

Data Index: 190590 (C4 (Web Text))

Max Activating Token Index: 778

Text #10

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3570. Min Act: -0.0000

Data Index: 299450 (C4 (Web Text))

Max Activating Token Index: 796

Text #11

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3424. Min Act: -0.0000

Data Index: 666529 (C4 (Web Text))

Max Activating Token Index: 344

Text #12

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.3601. Min Act: -0.0000

Data Index: 74842 (C4 (Web Text))

Max Activating Token Index: 851

Text #13

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.2959. Min Act: -0.0000

Data Index: 283583 (C4 (Web Text))

Max Activating Token Index: 280

Text #14

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.2678. Min Act: -0.0000

Data Index: 568082 (C4 (Web Text))

Max Activating Token Index: 870

Text #15

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.2766. Min Act: -0.0000

Data Index: 458366 (C4 (Web Text))

Max Activating Token Index: 215

Text #16

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.2611. Min Act: -0.0000

Data Index: 1415138 (Python Code)

Max Activating Token Index: 832

Text #17

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.2824. Min Act: -0.0000

Data Index: 202820 (C4 (Web Text))

Max Activating Token Index: 269

Text #18

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.2785. Min Act: -0.0000

Data Index: 1525658 (Python Code)

Max Activating Token Index: 532

Text #19

Max Range: 0.5110. Min Range: -0.5110

Max Act: 0.1983. Min Act: -0.0000

Data Index: 568264 (C4 (Web Text))

Max Activating Token Index: 944