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

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

Neuron 1056 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1839. Min Act: -0.0000

Data Index: 547552 (C4 (Web Text))

Max Activating Token Index: 745

Text #1

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1755. Min Act: -0.0000

Data Index: 517739 (C4 (Web Text))

Max Activating Token Index: 776

Text #2

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1555. Min Act: -0.0000

Data Index: 129827 (C4 (Web Text))

Max Activating Token Index: 863

Text #3

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1489. Min Act: -0.0000

Data Index: 404037 (C4 (Web Text))

Max Activating Token Index: 522

Text #4

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1415. Min Act: -0.0000

Data Index: 341177 (C4 (Web Text))

Max Activating Token Index: 114

Text #5

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1313. Min Act: -0.0001

Data Index: 1285214 (C4 (Web Text))

Max Activating Token Index: 589

Text #6

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1224. Min Act: -0.0001

Data Index: 587593 (C4 (Web Text))

Max Activating Token Index: 565

Text #7

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1213. Min Act: -0.0000

Data Index: 232326 (C4 (Web Text))

Max Activating Token Index: 779

Text #8

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1159. Min Act: -0.0001

Data Index: 1095226 (C4 (Web Text))

Max Activating Token Index: 778

Text #9

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1074. Min Act: -0.0000

Data Index: 1055763 (C4 (Web Text))

Max Activating Token Index: 855

Text #10

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1154. Min Act: -0.0000

Data Index: 1410576 (Python Code)

Max Activating Token Index: 362

Text #11

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1199. Min Act: -0.0001

Data Index: 945908 (C4 (Web Text))

Max Activating Token Index: 295

Text #12

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1156. Min Act: -0.0000

Data Index: 1146985 (C4 (Web Text))

Max Activating Token Index: 286

Text #13

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1113. Min Act: -0.0000

Data Index: 1517944 (Python Code)

Max Activating Token Index: 193

Text #14

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1148. Min Act: -0.0001

Data Index: 1056541 (C4 (Web Text))

Max Activating Token Index: 369

Text #15

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1020. Min Act: -0.0001

Data Index: 14741 (C4 (Web Text))

Max Activating Token Index: 905

Text #16

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1078. Min Act: -0.0000

Data Index: 1274263 (C4 (Web Text))

Max Activating Token Index: 928

Text #17

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1203. Min Act: -0.0000

Data Index: 1458610 (Python Code)

Max Activating Token Index: 605

Text #18

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1146. Min Act: -0.0000

Data Index: 788901 (C4 (Web Text))

Max Activating Token Index: 602

Text #19

Max Range: 0.1839. Min Range: -0.1839

Max Act: 0.1088. Min Act: -0.0000

Data Index: 1159299 (C4 (Web Text))

Max Activating Token Index: 392