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

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

Neuron 952 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1804. Min Act: -0.0001

Data Index: 1319861 (C4 (Web Text))

Max Activating Token Index: 1002

Text #1

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1727. Min Act: -0.0001

Data Index: 849140 (C4 (Web Text))

Max Activating Token Index: 636

Text #2

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1350. Min Act: -0.0001

Data Index: 495221 (C4 (Web Text))

Max Activating Token Index: 867

Text #3

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1484. Min Act: -0.0001

Data Index: 676440 (C4 (Web Text))

Max Activating Token Index: 245

Text #4

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1582. Min Act: -0.0001

Data Index: 1383488 (Python Code)

Max Activating Token Index: 349

Text #5

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1579. Min Act: -0.0001

Data Index: 311159 (C4 (Web Text))

Max Activating Token Index: 70

Text #6

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1388. Min Act: -0.0001

Data Index: 1075240 (C4 (Web Text))

Max Activating Token Index: 526

Text #7

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1254. Min Act: -0.0001

Data Index: 1135419 (C4 (Web Text))

Max Activating Token Index: 755

Text #8

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1138. Min Act: -0.0001

Data Index: 883421 (C4 (Web Text))

Max Activating Token Index: 667

Text #9

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1252. Min Act: -0.0001

Data Index: 104619 (C4 (Web Text))

Max Activating Token Index: 906

Text #10

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1236. Min Act: -0.0001

Data Index: 697537 (C4 (Web Text))

Max Activating Token Index: 1002

Text #11

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1108. Min Act: -0.0001

Data Index: 1162028 (C4 (Web Text))

Max Activating Token Index: 891

Text #12

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1186. Min Act: -0.0001

Data Index: 1369447 (Python Code)

Max Activating Token Index: 911

Text #13

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1111. Min Act: -0.0001

Data Index: 1582907 (Python Code)

Max Activating Token Index: 33

Text #14

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1142. Min Act: -0.0001

Data Index: 205386 (C4 (Web Text))

Max Activating Token Index: 165

Text #15

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1103. Min Act: -0.0001

Data Index: 1263954 (C4 (Web Text))

Max Activating Token Index: 187

Text #16

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1005. Min Act: -0.0001

Data Index: 1206762 (C4 (Web Text))

Max Activating Token Index: 1003

Text #17

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.0991. Min Act: -0.0001

Data Index: 911964 (C4 (Web Text))

Max Activating Token Index: 738

Text #18

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1051. Min Act: -0.0001

Data Index: 86137 (C4 (Web Text))

Max Activating Token Index: 258

Text #19

Max Range: 0.1804. Min Range: -0.1804

Max Act: 0.1059. Min Act: -0.0001

Data Index: 987872 (C4 (Web Text))

Max Activating Token Index: 902