Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2717 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.8462. Min Act: -0.0003

Data Index: 1350957 (C4 (Web Text))

Max Activating Token Index: 278

Text #1

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.9040. Min Act: -0.0003

Data Index: 804890 (C4 (Web Text))

Max Activating Token Index: 278

Text #2

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.8299. Min Act: -0.0003

Data Index: 574777 (C4 (Web Text))

Max Activating Token Index: 292

Text #3

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7908. Min Act: -0.0003

Data Index: 900979 (C4 (Web Text))

Max Activating Token Index: 424

Text #4

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7331. Min Act: -0.0003

Data Index: 607720 (C4 (Web Text))

Max Activating Token Index: 529

Text #5

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7319. Min Act: -0.0003

Data Index: 1358794 (C4 (Web Text))

Max Activating Token Index: 499

Text #6

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7999. Min Act: -0.0003

Data Index: 1347425 (C4 (Web Text))

Max Activating Token Index: 432

Text #7

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7265. Min Act: -0.0003

Data Index: 105920 (C4 (Web Text))

Max Activating Token Index: 713

Text #8

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7222. Min Act: -0.0003

Data Index: 1056640 (C4 (Web Text))

Max Activating Token Index: 152

Text #9

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.8090. Min Act: -0.0003

Data Index: 160976 (C4 (Web Text))

Max Activating Token Index: 252

Text #10

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7585. Min Act: -0.0003

Data Index: 403521 (C4 (Web Text))

Max Activating Token Index: 676

Text #11

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.8615. Min Act: -0.0003

Data Index: 805732 (C4 (Web Text))

Max Activating Token Index: 441

Text #12

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7254. Min Act: -0.0003

Data Index: 910277 (C4 (Web Text))

Max Activating Token Index: 812

Text #13

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7405. Min Act: -0.0003

Data Index: 755734 (C4 (Web Text))

Max Activating Token Index: 363

Text #14

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7199. Min Act: -0.0003

Data Index: 631319 (C4 (Web Text))

Max Activating Token Index: 220

Text #15

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.6984. Min Act: -0.0003

Data Index: 264684 (C4 (Web Text))

Max Activating Token Index: 115

Text #16

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.6329. Min Act: -0.0003

Data Index: 1058110 (C4 (Web Text))

Max Activating Token Index: 667

Text #17

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7074. Min Act: -0.0003

Data Index: 351963 (C4 (Web Text))

Max Activating Token Index: 806

Text #18

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.7983. Min Act: -0.0003

Data Index: 383258 (C4 (Web Text))

Max Activating Token Index: 306

Text #19

Max Range: 0.9040. Min Range: -0.9040

Max Act: 0.6357. Min Act: -0.0003

Data Index: 396530 (C4 (Web Text))

Max Activating Token Index: 517