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

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

Neuron 1591 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.8367. Min Act: -0.0000

Data Index: 680185 (C4 (Web Text))

Max Activating Token Index: 413

Text #1

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.7648. Min Act: -0.0000

Data Index: 409912 (C4 (Web Text))

Max Activating Token Index: 807

Text #2

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.7596. Min Act: -0.0000

Data Index: 949746 (C4 (Web Text))

Max Activating Token Index: 177

Text #3

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.7116. Min Act: -0.0000

Data Index: 182192 (C4 (Web Text))

Max Activating Token Index: 954

Text #4

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.7320. Min Act: -0.0000

Data Index: 199616 (C4 (Web Text))

Max Activating Token Index: 797

Text #5

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6802. Min Act: -0.0000

Data Index: 315786 (C4 (Web Text))

Max Activating Token Index: 763

Text #6

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6793. Min Act: -0.0000

Data Index: 289997 (C4 (Web Text))

Max Activating Token Index: 300

Text #7

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6849. Min Act: -0.0000

Data Index: 1437140 (Python Code)

Max Activating Token Index: 644

Text #8

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6621. Min Act: -0.0000

Data Index: 478674 (C4 (Web Text))

Max Activating Token Index: 89

Text #9

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6232. Min Act: -0.0000

Data Index: 1251989 (C4 (Web Text))

Max Activating Token Index: 516

Text #10

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6327. Min Act: -0.0000

Data Index: 1219416 (C4 (Web Text))

Max Activating Token Index: 413

Text #11

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6496. Min Act: -0.0000

Data Index: 981940 (C4 (Web Text))

Max Activating Token Index: 306

Text #12

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6394. Min Act: -0.0000

Data Index: 1354353 (C4 (Web Text))

Max Activating Token Index: 318

Text #13

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.5747. Min Act: -0.0000

Data Index: 677343 (C4 (Web Text))

Max Activating Token Index: 215

Text #14

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6237. Min Act: -0.0000

Data Index: 1165244 (C4 (Web Text))

Max Activating Token Index: 408

Text #15

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6010. Min Act: -0.0000

Data Index: 32008 (C4 (Web Text))

Max Activating Token Index: 926

Text #16

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.5765. Min Act: -0.0000

Data Index: 159001 (C4 (Web Text))

Max Activating Token Index: 436

Text #17

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.6271. Min Act: -0.0000

Data Index: 82944 (C4 (Web Text))

Max Activating Token Index: 905

Text #18

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.5728. Min Act: -0.0000

Data Index: 1504780 (Python Code)

Max Activating Token Index: 996

Text #19

Max Range: 0.8367. Min Range: -0.8367

Max Act: 0.5831. Min Act: -0.0000

Data Index: 251621 (C4 (Web Text))

Max Activating Token Index: 757