Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1837 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1753. Min Act: -0.0001

Data Index: 1134766 (C4 (Web Text))

Max Activating Token Index: 266

Text #1

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1621. Min Act: -0.0001

Data Index: 1646285 (Python Code)

Max Activating Token Index: 627

Text #2

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1366. Min Act: -0.0001

Data Index: 522110 (C4 (Web Text))

Max Activating Token Index: 62

Text #3

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1362. Min Act: -0.0001

Data Index: 549177 (C4 (Web Text))

Max Activating Token Index: 571

Text #4

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1436. Min Act: -0.0001

Data Index: 299916 (C4 (Web Text))

Max Activating Token Index: 154

Text #5

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1236. Min Act: -0.0001

Data Index: 1434303 (Python Code)

Max Activating Token Index: 42

Text #6

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1412. Min Act: -0.0001

Data Index: 757598 (C4 (Web Text))

Max Activating Token Index: 320

Text #7

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1289. Min Act: -0.0001

Data Index: 1468805 (Python Code)

Max Activating Token Index: 375

Text #8

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1227. Min Act: -0.0001

Data Index: 311480 (C4 (Web Text))

Max Activating Token Index: 722

Text #9

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1166. Min Act: -0.0001

Data Index: 159834 (C4 (Web Text))

Max Activating Token Index: 132

Text #10

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1237. Min Act: -0.0001

Data Index: 1615649 (Python Code)

Max Activating Token Index: 952

Text #11

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1143. Min Act: -0.0001

Data Index: 1557760 (Python Code)

Max Activating Token Index: 951

Text #12

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1167. Min Act: -0.0001

Data Index: 837059 (C4 (Web Text))

Max Activating Token Index: 932

Text #13

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1157. Min Act: -0.0001

Data Index: 1406044 (Python Code)

Max Activating Token Index: 29

Text #14

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1043. Min Act: -0.0001

Data Index: 85640 (C4 (Web Text))

Max Activating Token Index: 678

Text #15

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1037. Min Act: -0.0001

Data Index: 1351915 (C4 (Web Text))

Max Activating Token Index: 729

Text #16

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1039. Min Act: -0.0001

Data Index: 443403 (C4 (Web Text))

Max Activating Token Index: 901

Text #17

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1047. Min Act: -0.0001

Data Index: 530517 (C4 (Web Text))

Max Activating Token Index: 550

Text #18

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.1160. Min Act: -0.0001

Data Index: 1444745 (Python Code)

Max Activating Token Index: 336

Text #19

Max Range: 0.1753. Min Range: -0.1753

Max Act: 0.0979. Min Act: -0.0001

Data Index: 106731 (C4 (Web Text))

Max Activating Token Index: 278