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

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

Neuron 1400 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1525. Min Act: -0.0000

Data Index: 735976 (C4 (Web Text))

Max Activating Token Index: 771

Text #1

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1580. Min Act: -0.0000

Data Index: 426278 (C4 (Web Text))

Max Activating Token Index: 913

Text #2

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1346. Min Act: -0.0000

Data Index: 961150 (C4 (Web Text))

Max Activating Token Index: 779

Text #3

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1437. Min Act: -0.0000

Data Index: 605617 (C4 (Web Text))

Max Activating Token Index: 950

Text #4

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1349. Min Act: -0.0000

Data Index: 450127 (C4 (Web Text))

Max Activating Token Index: 910

Text #5

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1287. Min Act: -0.0000

Data Index: 716389 (C4 (Web Text))

Max Activating Token Index: 728

Text #6

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1233. Min Act: -0.0000

Data Index: 1351612 (C4 (Web Text))

Max Activating Token Index: 580

Text #7

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1225. Min Act: -0.0000

Data Index: 66557 (C4 (Web Text))

Max Activating Token Index: 732

Text #8

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1258. Min Act: -0.0000

Data Index: 308583 (C4 (Web Text))

Max Activating Token Index: 518

Text #9

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1271. Min Act: -0.0000

Data Index: 790237 (C4 (Web Text))

Max Activating Token Index: 356

Text #10

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1071. Min Act: -0.0000

Data Index: 175545 (C4 (Web Text))

Max Activating Token Index: 1022

Text #11

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1096. Min Act: -0.0000

Data Index: 1199241 (C4 (Web Text))

Max Activating Token Index: 390

Text #12

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1077. Min Act: -0.0000

Data Index: 1245201 (C4 (Web Text))

Max Activating Token Index: 548

Text #13

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1090. Min Act: -0.0000

Data Index: 795133 (C4 (Web Text))

Max Activating Token Index: 491

Text #14

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1163. Min Act: -0.0000

Data Index: 1545323 (Python Code)

Max Activating Token Index: 949

Text #15

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1053. Min Act: -0.0000

Data Index: 949358 (C4 (Web Text))

Max Activating Token Index: 554

Text #16

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1080. Min Act: -0.0000

Data Index: 1336997 (C4 (Web Text))

Max Activating Token Index: 571

Text #17

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.0998. Min Act: -0.0000

Data Index: 790787 (C4 (Web Text))

Max Activating Token Index: 953

Text #18

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.0999. Min Act: -0.0000

Data Index: 1108702 (C4 (Web Text))

Max Activating Token Index: 676

Text #19

Max Range: 0.1580. Min Range: -0.1580

Max Act: 0.1034. Min Act: -0.0000

Data Index: 196544 (C4 (Web Text))

Max Activating Token Index: 679