Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1965 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1299. Min Act: -0.0000

Data Index: 1426536 (Python Code)

Max Activating Token Index: 369

Text #1

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1375. Min Act: -0.0000

Data Index: 128284 (C4 (Web Text))

Max Activating Token Index: 857

Text #2

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1338. Min Act: -0.0000

Data Index: 469814 (C4 (Web Text))

Max Activating Token Index: 760

Text #3

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1514. Min Act: -0.0000

Data Index: 613965 (C4 (Web Text))

Max Activating Token Index: 436

Text #4

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1275. Min Act: -0.0000

Data Index: 1534176 (Python Code)

Max Activating Token Index: 952

Text #5

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1150. Min Act: -0.0000

Data Index: 325285 (C4 (Web Text))

Max Activating Token Index: 892

Text #6

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1190. Min Act: -0.0000

Data Index: 1384399 (Python Code)

Max Activating Token Index: 191

Text #7

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1166. Min Act: -0.0000

Data Index: 1437002 (Python Code)

Max Activating Token Index: 676

Text #8

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1144. Min Act: -0.0000

Data Index: 1461850 (Python Code)

Max Activating Token Index: 129

Text #9

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1182. Min Act: -0.0000

Data Index: 1524023 (Python Code)

Max Activating Token Index: 121

Text #10

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1117. Min Act: -0.0000

Data Index: 1118334 (C4 (Web Text))

Max Activating Token Index: 745

Text #11

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1105. Min Act: -0.0000

Data Index: 586987 (C4 (Web Text))

Max Activating Token Index: 479

Text #12

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1062. Min Act: -0.0000

Data Index: 349381 (C4 (Web Text))

Max Activating Token Index: 896

Text #13

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1014. Min Act: -0.0000

Data Index: 1530655 (Python Code)

Max Activating Token Index: 100

Text #14

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1135. Min Act: -0.0000

Data Index: 456394 (C4 (Web Text))

Max Activating Token Index: 530

Text #15

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1037. Min Act: -0.0000

Data Index: 647245 (C4 (Web Text))

Max Activating Token Index: 390

Text #16

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1028. Min Act: -0.0000

Data Index: 314108 (C4 (Web Text))

Max Activating Token Index: 279

Text #17

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.1049. Min Act: -0.0000

Data Index: 1370800 (Python Code)

Max Activating Token Index: 739

Text #18

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.0955. Min Act: -0.0000

Data Index: 1172544 (C4 (Web Text))

Max Activating Token Index: 873

Text #19

Max Range: 0.1514. Min Range: -0.1514

Max Act: 0.0985. Min Act: -0.0000

Data Index: 269599 (C4 (Web Text))

Max Activating Token Index: 434