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

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

Neuron 2528 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: 0.1448. Min Range: -0.1448

Max Act: 0.1448. Min Act: -0.0001

Data Index: 44000 (C4 (Web Text))

Max Activating Token Index: 156

Text #1

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1380. Min Act: -0.0001

Data Index: 788931 (C4 (Web Text))

Max Activating Token Index: 485

Text #2

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1285. Min Act: -0.0001

Data Index: 613841 (C4 (Web Text))

Max Activating Token Index: 573

Text #3

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1292. Min Act: -0.0001

Data Index: 49000 (C4 (Web Text))

Max Activating Token Index: 82

Text #4

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1249. Min Act: -0.0001

Data Index: 1202731 (C4 (Web Text))

Max Activating Token Index: 733

Text #5

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1267. Min Act: -0.0001

Data Index: 796188 (C4 (Web Text))

Max Activating Token Index: 699

Text #6

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1211. Min Act: -0.0001

Data Index: 1534950 (Python Code)

Max Activating Token Index: 850

Text #7

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1235. Min Act: -0.0001

Data Index: 256481 (C4 (Web Text))

Max Activating Token Index: 53

Text #8

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1293. Min Act: -0.0001

Data Index: 463543 (C4 (Web Text))

Max Activating Token Index: 599

Text #9

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1134. Min Act: -0.0001

Data Index: 130372 (C4 (Web Text))

Max Activating Token Index: 68

Text #10

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1195. Min Act: -0.0001

Data Index: 1608808 (Python Code)

Max Activating Token Index: 340

Text #11

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1278. Min Act: -0.0001

Data Index: 841471 (C4 (Web Text))

Max Activating Token Index: 83

Text #12

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1139. Min Act: -0.0001

Data Index: 816710 (C4 (Web Text))

Max Activating Token Index: 735

Text #13

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1150. Min Act: -0.0001

Data Index: 93348 (C4 (Web Text))

Max Activating Token Index: 662

Text #14

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1123. Min Act: -0.0001

Data Index: 1404854 (Python Code)

Max Activating Token Index: 61

Text #15

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1045. Min Act: -0.0001

Data Index: 1650780 (Python Code)

Max Activating Token Index: 598

Text #16

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1095. Min Act: -0.0001

Data Index: 1388823 (Python Code)

Max Activating Token Index: 5

Text #17

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.0964. Min Act: -0.0001

Data Index: 697677 (C4 (Web Text))

Max Activating Token Index: 35

Text #18

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1132. Min Act: -0.0001

Data Index: 195284 (C4 (Web Text))

Max Activating Token Index: 946

Text #19

Max Range: 0.1448. Min Range: -0.1448

Max Act: 0.1087. Min Act: -0.0001

Data Index: 1589885 (Python Code)

Max Activating Token Index: 610