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

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

Neuron 3640 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: 0.0236. Min Range: -0.0236

Max Act: 0.0236. Min Act: -0.0001

Data Index: 458984 (C4 (Web Text))

Max Activating Token Index: 567

Text #1

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0198. Min Act: -0.0001

Data Index: 165928 (C4 (Web Text))

Max Activating Token Index: 750

Text #2

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0165. Min Act: -0.0001

Data Index: 968797 (C4 (Web Text))

Max Activating Token Index: 346

Text #3

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0204. Min Act: -0.0001

Data Index: 900397 (C4 (Web Text))

Max Activating Token Index: 1014

Text #4

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0194. Min Act: -0.0001

Data Index: 453598 (C4 (Web Text))

Max Activating Token Index: 996

Text #5

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0177. Min Act: -0.0001

Data Index: 1442707 (Python Code)

Max Activating Token Index: 399

Text #6

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0156. Min Act: -0.0001

Data Index: 658152 (C4 (Web Text))

Max Activating Token Index: 666

Text #7

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0161. Min Act: -0.0001

Data Index: 210622 (C4 (Web Text))

Max Activating Token Index: 462

Text #8

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0170. Min Act: -0.0001

Data Index: 871061 (C4 (Web Text))

Max Activating Token Index: 291

Text #9

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0151. Min Act: -0.0001

Data Index: 87257 (C4 (Web Text))

Max Activating Token Index: 263

Text #10

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0155. Min Act: -0.0001

Data Index: 1253712 (C4 (Web Text))

Max Activating Token Index: 881

Text #11

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0141. Min Act: -0.0001

Data Index: 1485593 (Python Code)

Max Activating Token Index: 906

Text #12

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0149. Min Act: -0.0001

Data Index: 1535558 (Python Code)

Max Activating Token Index: 651

Text #13

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0158. Min Act: -0.0001

Data Index: 618789 (C4 (Web Text))

Max Activating Token Index: 767

Text #14

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0144. Min Act: -0.0001

Data Index: 1580582 (Python Code)

Max Activating Token Index: 649

Text #15

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0151. Min Act: -0.0001

Data Index: 1524195 (Python Code)

Max Activating Token Index: 293

Text #16

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0149. Min Act: -0.0001

Data Index: 42749 (C4 (Web Text))

Max Activating Token Index: 758

Text #17

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0139. Min Act: -0.0001

Data Index: 1107380 (C4 (Web Text))

Max Activating Token Index: 284

Text #18

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0152. Min Act: -0.0001

Data Index: 1413734 (Python Code)

Max Activating Token Index: 447

Text #19

Max Range: 0.0236. Min Range: -0.0236

Max Act: 0.0132. Min Act: -0.0001

Data Index: 241023 (C4 (Web Text))

Max Activating Token Index: 692