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

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

Neuron 3218 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0381. Min Range: -1.0381

Max Act: 1.0381. Min Act: -0.0001

Data Index: 1096240 (C4 (Web Text))

Max Activating Token Index: 6

Text #1

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.9108. Min Act: -0.0001

Data Index: 570242 (C4 (Web Text))

Max Activating Token Index: 49

Text #2

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.8761. Min Act: -0.0001

Data Index: 745065 (C4 (Web Text))

Max Activating Token Index: 960

Text #3

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.9057. Min Act: -0.0001

Data Index: 1073382 (C4 (Web Text))

Max Activating Token Index: 306

Text #4

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.8205. Min Act: -0.0001

Data Index: 206092 (C4 (Web Text))

Max Activating Token Index: 122

Text #5

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.8385. Min Act: -0.0001

Data Index: 277988 (C4 (Web Text))

Max Activating Token Index: 372

Text #6

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7878. Min Act: -0.0001

Data Index: 1543957 (Python Code)

Max Activating Token Index: 351

Text #7

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7736. Min Act: -0.0001

Data Index: 505787 (C4 (Web Text))

Max Activating Token Index: 25

Text #8

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7623. Min Act: -0.0001

Data Index: 996028 (C4 (Web Text))

Max Activating Token Index: 460

Text #9

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.8139. Min Act: -0.0001

Data Index: 1303420 (C4 (Web Text))

Max Activating Token Index: 779

Text #10

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7577. Min Act: -0.0001

Data Index: 1527344 (Python Code)

Max Activating Token Index: 172

Text #11

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7833. Min Act: -0.0001

Data Index: 265017 (C4 (Web Text))

Max Activating Token Index: 272

Text #12

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7720. Min Act: -0.0001

Data Index: 422312 (C4 (Web Text))

Max Activating Token Index: 820

Text #13

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7397. Min Act: -0.0001

Data Index: 908023 (C4 (Web Text))

Max Activating Token Index: 856

Text #14

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7364. Min Act: -0.0001

Data Index: 688433 (C4 (Web Text))

Max Activating Token Index: 316

Text #15

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7349. Min Act: -0.0001

Data Index: 388424 (C4 (Web Text))

Max Activating Token Index: 118

Text #16

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.6673. Min Act: -0.0001

Data Index: 600530 (C4 (Web Text))

Max Activating Token Index: 997

Text #17

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.6828. Min Act: -0.0001

Data Index: 1613502 (Python Code)

Max Activating Token Index: 258

Text #18

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7144. Min Act: -0.0001

Data Index: 804028 (C4 (Web Text))

Max Activating Token Index: 86

Text #19

Max Range: 1.0381. Min Range: -1.0381

Max Act: 0.7478. Min Act: -0.0001

Data Index: 567569 (C4 (Web Text))

Max Activating Token Index: 1014