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

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

Neuron 4512 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5889. Min Range: -2.5889

Max Act: 2.5716. Min Act: -0.0000

Data Index: 854218 (C4 (Web Text))

Max Activating Token Index: 930

Text #1

Max Range: 2.5889. Min Range: -2.5889

Max Act: 2.5889. Min Act: -0.0000

Data Index: 1274502 (C4 (Web Text))

Max Activating Token Index: 638

Text #2

Max Range: 2.5889. Min Range: -2.5889

Max Act: 2.0593. Min Act: -0.0000

Data Index: 1337696 (C4 (Web Text))

Max Activating Token Index: 870

Text #3

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.7238. Min Act: -0.0000

Data Index: 1059047 (C4 (Web Text))

Max Activating Token Index: 626

Text #4

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.6452. Min Act: -0.0000

Data Index: 475425 (C4 (Web Text))

Max Activating Token Index: 597

Text #5

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.5146. Min Act: -0.0000

Data Index: 584957 (C4 (Web Text))

Max Activating Token Index: 324

Text #6

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.3915. Min Act: -0.0000

Data Index: 597093 (C4 (Web Text))

Max Activating Token Index: 682

Text #7

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.5126. Min Act: -0.0000

Data Index: 833926 (C4 (Web Text))

Max Activating Token Index: 224

Text #8

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.3947. Min Act: -0.0000

Data Index: 163568 (C4 (Web Text))

Max Activating Token Index: 516

Text #9

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.4046. Min Act: -0.0000

Data Index: 1023414 (C4 (Web Text))

Max Activating Token Index: 565

Text #10

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.3457. Min Act: -0.0000

Data Index: 1531366 (Python Code)

Max Activating Token Index: 823

Text #11

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.2848. Min Act: -0.0000

Data Index: 1394138 (Python Code)

Max Activating Token Index: 907

Text #12

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.2649. Min Act: -0.0000

Data Index: 1473010 (Python Code)

Max Activating Token Index: 146

Text #13

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.1522. Min Act: -0.0000

Data Index: 1641177 (Python Code)

Max Activating Token Index: 828

Text #14

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.1824. Min Act: -0.0000

Data Index: 1224232 (C4 (Web Text))

Max Activating Token Index: 204

Text #15

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.2297. Min Act: -0.0000

Data Index: 51009 (C4 (Web Text))

Max Activating Token Index: 821

Text #16

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.2052. Min Act: -0.0000

Data Index: 88685 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.0963. Min Act: -0.0000

Data Index: 1367394 (Python Code)

Max Activating Token Index: 411

Text #18

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.1025. Min Act: -0.0000

Data Index: 119754 (C4 (Web Text))

Max Activating Token Index: 616

Text #19

Max Range: 2.5889. Min Range: -2.5889

Max Act: 1.1514. Min Act: -0.0000

Data Index: 709991 (C4 (Web Text))

Max Activating Token Index: 32