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

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

Neuron 655 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: 1.5344. Min Range: -1.5344

Max Act: 1.5344. Min Act: -0.0000

Data Index: 492068 (C4 (Web Text))

Max Activating Token Index: 586

Text #1

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.2987. Min Act: -0.0000

Data Index: 622932 (C4 (Web Text))

Max Activating Token Index: 435

Text #2

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.1314. Min Act: -0.0000

Data Index: 634433 (C4 (Web Text))

Max Activating Token Index: 646

Text #3

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.1913. Min Act: -0.0000

Data Index: 888971 (C4 (Web Text))

Max Activating Token Index: 794

Text #4

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.1031. Min Act: -0.0000

Data Index: 641232 (C4 (Web Text))

Max Activating Token Index: 584

Text #5

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.0817. Min Act: -0.0000

Data Index: 1220681 (C4 (Web Text))

Max Activating Token Index: 93

Text #6

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.0670. Min Act: -0.0000

Data Index: 1091844 (C4 (Web Text))

Max Activating Token Index: 224

Text #7

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9376. Min Act: -0.0000

Data Index: 1274922 (C4 (Web Text))

Max Activating Token Index: 56

Text #8

Max Range: 1.5344. Min Range: -1.5344

Max Act: 1.0351. Min Act: -0.0000

Data Index: 544952 (C4 (Web Text))

Max Activating Token Index: 589

Text #9

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9624. Min Act: -0.0000

Data Index: 29588 (C4 (Web Text))

Max Activating Token Index: 609

Text #10

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9790. Min Act: -0.0000

Data Index: 17007 (C4 (Web Text))

Max Activating Token Index: 885

Text #11

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9483. Min Act: -0.0000

Data Index: 383735 (C4 (Web Text))

Max Activating Token Index: 448

Text #12

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9541. Min Act: -0.0000

Data Index: 106326 (C4 (Web Text))

Max Activating Token Index: 82

Text #13

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9482. Min Act: -0.0000

Data Index: 179531 (C4 (Web Text))

Max Activating Token Index: 218

Text #14

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9744. Min Act: -0.0000

Data Index: 96328 (C4 (Web Text))

Max Activating Token Index: 110

Text #15

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9116. Min Act: -0.0000

Data Index: 269194 (C4 (Web Text))

Max Activating Token Index: 394

Text #16

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9340. Min Act: -0.0000

Data Index: 26795 (C4 (Web Text))

Max Activating Token Index: 495

Text #17

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.8631. Min Act: -0.0000

Data Index: 1541608 (Python Code)

Max Activating Token Index: 259

Text #18

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.8698. Min Act: -0.0000

Data Index: 803458 (C4 (Web Text))

Max Activating Token Index: 178

Text #19

Max Range: 1.5344. Min Range: -1.5344

Max Act: 0.9294. Min Act: -0.0000

Data Index: 210038 (C4 (Web Text))

Max Activating Token Index: 931