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

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

Neuron 3992 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.8192. Min Act: -0.0001

Data Index: 587410 (C4 (Web Text))

Max Activating Token Index: 565

Text #1

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6815. Min Act: -0.0001

Data Index: 940330 (C4 (Web Text))

Max Activating Token Index: 173

Text #2

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.7772. Min Act: -0.0001

Data Index: 728329 (C4 (Web Text))

Max Activating Token Index: 81

Text #3

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6670. Min Act: -0.0001

Data Index: 470175 (C4 (Web Text))

Max Activating Token Index: 460

Text #4

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6515. Min Act: -0.0001

Data Index: 674801 (C4 (Web Text))

Max Activating Token Index: 532

Text #5

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.7167. Min Act: -0.0001

Data Index: 945185 (C4 (Web Text))

Max Activating Token Index: 298

Text #6

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.5710. Min Act: -0.0001

Data Index: 277559 (C4 (Web Text))

Max Activating Token Index: 724

Text #7

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6399. Min Act: -0.0001

Data Index: 652203 (C4 (Web Text))

Max Activating Token Index: 39

Text #8

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6251. Min Act: -0.0001

Data Index: 542157 (C4 (Web Text))

Max Activating Token Index: 27

Text #9

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6176. Min Act: -0.0001

Data Index: 457166 (C4 (Web Text))

Max Activating Token Index: 483

Text #10

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6478. Min Act: -0.0001

Data Index: 429184 (C4 (Web Text))

Max Activating Token Index: 528

Text #11

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6083. Min Act: -0.0001

Data Index: 1192260 (C4 (Web Text))

Max Activating Token Index: 494

Text #12

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.6269. Min Act: -0.0001

Data Index: 450260 (C4 (Web Text))

Max Activating Token Index: 896

Text #13

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.5928. Min Act: -0.0000

Data Index: 1549921 (Python Code)

Max Activating Token Index: 532

Text #14

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.5511. Min Act: -0.0001

Data Index: 1229499 (C4 (Web Text))

Max Activating Token Index: 652

Text #15

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.5867. Min Act: -0.0001

Data Index: 143706 (C4 (Web Text))

Max Activating Token Index: 848

Text #16

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.5335. Min Act: -0.0001

Data Index: 1328303 (C4 (Web Text))

Max Activating Token Index: 364

Text #17

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.5266. Min Act: -0.0001

Data Index: 507725 (C4 (Web Text))

Max Activating Token Index: 35

Text #18

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.4874. Min Act: -0.0001

Data Index: 888721 (C4 (Web Text))

Max Activating Token Index: 633

Text #19

Max Range: 0.8192. Min Range: -0.8192

Max Act: 0.3990. Min Act: -0.0001

Data Index: 1347560 (C4 (Web Text))

Max Activating Token Index: 791