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

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

Neuron 786 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: 0.9668. Min Range: -0.9668

Max Act: 0.9454. Min Act: -0.0001

Data Index: 595566 (C4 (Web Text))

Max Activating Token Index: 875

Text #1

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.9668. Min Act: -0.0001

Data Index: 217553 (C4 (Web Text))

Max Activating Token Index: 535

Text #2

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.9102. Min Act: -0.0001

Data Index: 198360 (C4 (Web Text))

Max Activating Token Index: 991

Text #3

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8917. Min Act: -0.0001

Data Index: 400679 (C4 (Web Text))

Max Activating Token Index: 447

Text #4

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.9143. Min Act: -0.0001

Data Index: 270714 (C4 (Web Text))

Max Activating Token Index: 114

Text #5

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8813. Min Act: -0.0001

Data Index: 299785 (C4 (Web Text))

Max Activating Token Index: 965

Text #6

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8687. Min Act: -0.0001

Data Index: 229707 (C4 (Web Text))

Max Activating Token Index: 53

Text #7

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8711. Min Act: -0.0001

Data Index: 1014175 (C4 (Web Text))

Max Activating Token Index: 20

Text #8

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8580. Min Act: -0.0001

Data Index: 426149 (C4 (Web Text))

Max Activating Token Index: 986

Text #9

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8800. Min Act: -0.0001

Data Index: 647439 (C4 (Web Text))

Max Activating Token Index: 10

Text #10

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8992. Min Act: -0.0001

Data Index: 788259 (C4 (Web Text))

Max Activating Token Index: 749

Text #11

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8179. Min Act: -0.0001

Data Index: 768832 (C4 (Web Text))

Max Activating Token Index: 31

Text #12

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8138. Min Act: -0.0001

Data Index: 149312 (C4 (Web Text))

Max Activating Token Index: 150

Text #13

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8094. Min Act: -0.0001

Data Index: 1034143 (C4 (Web Text))

Max Activating Token Index: 696

Text #14

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8830. Min Act: -0.0001

Data Index: 1024124 (C4 (Web Text))

Max Activating Token Index: 522

Text #15

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8166. Min Act: -0.0001

Data Index: 1533848 (Python Code)

Max Activating Token Index: 87

Text #16

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8235. Min Act: -0.0001

Data Index: 1059329 (C4 (Web Text))

Max Activating Token Index: 489

Text #17

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.7927. Min Act: -0.0001

Data Index: 805231 (C4 (Web Text))

Max Activating Token Index: 397

Text #18

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8705. Min Act: -0.0001

Data Index: 674502 (C4 (Web Text))

Max Activating Token Index: 336

Text #19

Max Range: 0.9668. Min Range: -0.9668

Max Act: 0.8197. Min Act: -0.0001

Data Index: 262828 (C4 (Web Text))

Max Activating Token Index: 732