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

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

Neuron 4810 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5927. Min Range: -2.5927

Max Act: 2.5927. Min Act: -0.0000

Data Index: 1189194 (C4 (Web Text))

Max Activating Token Index: 84

Text #1

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.9735. Min Act: -0.0000

Data Index: 423728 (C4 (Web Text))

Max Activating Token Index: 209

Text #2

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.9102. Min Act: -0.0000

Data Index: 510328 (C4 (Web Text))

Max Activating Token Index: 268

Text #3

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.7629. Min Act: -0.0000

Data Index: 660986 (C4 (Web Text))

Max Activating Token Index: 261

Text #4

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.8450. Min Act: -0.0000

Data Index: 929915 (C4 (Web Text))

Max Activating Token Index: 422

Text #5

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.7423. Min Act: -0.0000

Data Index: 1447160 (Python Code)

Max Activating Token Index: 30

Text #6

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.6584. Min Act: -0.0000

Data Index: 1369077 (Python Code)

Max Activating Token Index: 46

Text #7

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.7318. Min Act: -0.0000

Data Index: 101391 (C4 (Web Text))

Max Activating Token Index: 28

Text #8

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.7076. Min Act: -0.0000

Data Index: 1183775 (C4 (Web Text))

Max Activating Token Index: 31

Text #9

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.6857. Min Act: -0.0000

Data Index: 517830 (C4 (Web Text))

Max Activating Token Index: 528

Text #10

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.7680. Min Act: -0.0000

Data Index: 442557 (C4 (Web Text))

Max Activating Token Index: 948

Text #11

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.6348. Min Act: -0.0000

Data Index: 1483359 (Python Code)

Max Activating Token Index: 169

Text #12

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.6266. Min Act: -0.0000

Data Index: 1500968 (Python Code)

Max Activating Token Index: 55

Text #13

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.6417. Min Act: -0.0000

Data Index: 1606694 (Python Code)

Max Activating Token Index: 61

Text #14

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.5505. Min Act: -0.0000

Data Index: 1474142 (Python Code)

Max Activating Token Index: 139

Text #15

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.7215. Min Act: -0.0000

Data Index: 1553423 (Python Code)

Max Activating Token Index: 37

Text #16

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.5427. Min Act: -0.0000

Data Index: 1651667 (Python Code)

Max Activating Token Index: 98

Text #17

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.5059. Min Act: -0.0000

Data Index: 1472752 (Python Code)

Max Activating Token Index: 190

Text #18

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.6373. Min Act: -0.0000

Data Index: 1363767 (Python Code)

Max Activating Token Index: 173

Text #19

Max Range: 2.5927. Min Range: -2.5927

Max Act: 1.5228. Min Act: -0.0000

Data Index: 1392599 (Python Code)

Max Activating Token Index: 85