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

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

Neuron 1620 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1317. Min Act: -0.0001

Data Index: 840079 (C4 (Web Text))

Max Activating Token Index: 104

Text #1

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1146. Min Act: -0.0001

Data Index: 538427 (C4 (Web Text))

Max Activating Token Index: 210

Text #2

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1332. Min Act: -0.0001

Data Index: 638726 (C4 (Web Text))

Max Activating Token Index: 980

Text #3

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1158. Min Act: -0.0002

Data Index: 3142 (C4 (Web Text))

Max Activating Token Index: 877

Text #4

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1261. Min Act: -0.0001

Data Index: 1037494 (C4 (Web Text))

Max Activating Token Index: 124

Text #5

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1209. Min Act: -0.0001

Data Index: 647794 (C4 (Web Text))

Max Activating Token Index: 316

Text #6

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1169. Min Act: -0.0001

Data Index: 66527 (C4 (Web Text))

Max Activating Token Index: 613

Text #7

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1094. Min Act: -0.0001

Data Index: 667785 (C4 (Web Text))

Max Activating Token Index: 766

Text #8

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1142. Min Act: -0.0001

Data Index: 486184 (C4 (Web Text))

Max Activating Token Index: 349

Text #9

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1163. Min Act: -0.0001

Data Index: 1571576 (Python Code)

Max Activating Token Index: 942

Text #10

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1123. Min Act: -0.0001

Data Index: 719937 (C4 (Web Text))

Max Activating Token Index: 211

Text #11

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1120. Min Act: -0.0001

Data Index: 918689 (C4 (Web Text))

Max Activating Token Index: 11

Text #12

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1067. Min Act: -0.0001

Data Index: 112745 (C4 (Web Text))

Max Activating Token Index: 729

Text #13

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1100. Min Act: -0.0001

Data Index: 572561 (C4 (Web Text))

Max Activating Token Index: 298

Text #14

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1134. Min Act: -0.0001

Data Index: 382651 (C4 (Web Text))

Max Activating Token Index: 371

Text #15

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1117. Min Act: -0.0001

Data Index: 1159346 (C4 (Web Text))

Max Activating Token Index: 63

Text #16

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1051. Min Act: -0.0001

Data Index: 3270 (C4 (Web Text))

Max Activating Token Index: 689

Text #17

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1056. Min Act: -0.0001

Data Index: 457314 (C4 (Web Text))

Max Activating Token Index: 51

Text #18

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1056. Min Act: -0.0001

Data Index: 1353300 (C4 (Web Text))

Max Activating Token Index: 980

Text #19

Max Range: 0.1332. Min Range: -0.1332

Max Act: 0.1006. Min Act: -0.0001

Data Index: 579495 (C4 (Web Text))

Max Activating Token Index: 105