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

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

Neuron 26 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.9210. Min Act: -0.0001

Data Index: 544798 (C4 (Web Text))

Max Activating Token Index: 198

Text #1

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.5813. Min Act: -0.0001

Data Index: 826669 (C4 (Web Text))

Max Activating Token Index: 780

Text #2

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.6390. Min Act: -0.0001

Data Index: 1141848 (C4 (Web Text))

Max Activating Token Index: 621

Text #3

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.5251. Min Act: -0.0001

Data Index: 709444 (C4 (Web Text))

Max Activating Token Index: 454

Text #4

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.4480. Min Act: -0.0001

Data Index: 1494341 (Python Code)

Max Activating Token Index: 9

Text #5

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.4343. Min Act: -0.0001

Data Index: 1506125 (Python Code)

Max Activating Token Index: 595

Text #6

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.4242. Min Act: -0.0001

Data Index: 688095 (C4 (Web Text))

Max Activating Token Index: 334

Text #7

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.3353. Min Act: -0.0001

Data Index: 340957 (C4 (Web Text))

Max Activating Token Index: 835

Text #8

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.3330. Min Act: -0.0001

Data Index: 162754 (C4 (Web Text))

Max Activating Token Index: 945

Text #9

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.4825. Min Act: -0.0001

Data Index: 482407 (C4 (Web Text))

Max Activating Token Index: 809

Text #10

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.3454. Min Act: -0.0001

Data Index: 291236 (C4 (Web Text))

Max Activating Token Index: 109

Text #11

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.3191. Min Act: -0.0001

Data Index: 104189 (C4 (Web Text))

Max Activating Token Index: 713

Text #12

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.3117. Min Act: -0.0001

Data Index: 1290946 (C4 (Web Text))

Max Activating Token Index: 455

Text #13

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.2581. Min Act: -0.0001

Data Index: 1584026 (Python Code)

Max Activating Token Index: 9

Text #14

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.3197. Min Act: -0.0001

Data Index: 1213708 (C4 (Web Text))

Max Activating Token Index: 652

Text #15

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.2783. Min Act: -0.0001

Data Index: 190914 (C4 (Web Text))

Max Activating Token Index: 344

Text #16

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.2508. Min Act: -0.0001

Data Index: 306928 (C4 (Web Text))

Max Activating Token Index: 1021

Text #17

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.2112. Min Act: -0.0001

Data Index: 1475854 (Python Code)

Max Activating Token Index: 983

Text #18

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.2764. Min Act: -0.0001

Data Index: 83697 (C4 (Web Text))

Max Activating Token Index: 65

Text #19

Max Range: 1.9210. Min Range: -1.9210

Max Act: 1.2237. Min Act: -0.0001

Data Index: 1612277 (Python Code)

Max Activating Token Index: 170