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

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

Neuron 1091 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: 5.2364. Min Range: -5.2364

Max Act: 5.2364. Min Act: -0.0001

Data Index: 160971 (C4 (Web Text))

Max Activating Token Index: 600

Text #1

Max Range: 5.2364. Min Range: -5.2364

Max Act: 4.7285. Min Act: -0.0001

Data Index: 755760 (C4 (Web Text))

Max Activating Token Index: 628

Text #2

Max Range: 5.2364. Min Range: -5.2364

Max Act: 4.2294. Min Act: -0.0001

Data Index: 964820 (C4 (Web Text))

Max Activating Token Index: 850

Text #3

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.8935. Min Act: -0.0001

Data Index: 747249 (C4 (Web Text))

Max Activating Token Index: 572

Text #4

Max Range: 5.2364. Min Range: -5.2364

Max Act: 4.1490. Min Act: -0.0001

Data Index: 1189420 (C4 (Web Text))

Max Activating Token Index: 258

Text #5

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.9077. Min Act: -0.0001

Data Index: 196194 (C4 (Web Text))

Max Activating Token Index: 908

Text #6

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.8886. Min Act: -0.0001

Data Index: 693308 (C4 (Web Text))

Max Activating Token Index: 937

Text #7

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.5833. Min Act: -0.0001

Data Index: 451850 (C4 (Web Text))

Max Activating Token Index: 977

Text #8

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.9274. Min Act: -0.0001

Data Index: 696399 (C4 (Web Text))

Max Activating Token Index: 868

Text #9

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.8731. Min Act: -0.0001

Data Index: 837694 (C4 (Web Text))

Max Activating Token Index: 881

Text #10

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.7132. Min Act: -0.0001

Data Index: 1196395 (C4 (Web Text))

Max Activating Token Index: 168

Text #11

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.6843. Min Act: -0.0001

Data Index: 613521 (C4 (Web Text))

Max Activating Token Index: 306

Text #12

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.6171. Min Act: -0.0001

Data Index: 281099 (C4 (Web Text))

Max Activating Token Index: 39

Text #13

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.4507. Min Act: -0.0001

Data Index: 1248282 (C4 (Web Text))

Max Activating Token Index: 358

Text #14

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.4160. Min Act: -0.0001

Data Index: 1420693 (Python Code)

Max Activating Token Index: 844

Text #15

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.2245. Min Act: -0.0001

Data Index: 65043 (C4 (Web Text))

Max Activating Token Index: 626

Text #16

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.5403. Min Act: -0.0001

Data Index: 1114510 (C4 (Web Text))

Max Activating Token Index: 131

Text #17

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.5304. Min Act: -0.0001

Data Index: 229140 (C4 (Web Text))

Max Activating Token Index: 131

Text #18

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.5121. Min Act: -0.0001

Data Index: 1210021 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 5.2364. Min Range: -5.2364

Max Act: 3.5157. Min Act: -0.0001

Data Index: 877450 (C4 (Web Text))

Max Activating Token Index: 613