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

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

Neuron 4853 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.7372. Min Act: -0.0000

Data Index: 10658 (C4 (Web Text))

Max Activating Token Index: 171

Text #1

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.7624. Min Act: -0.0000

Data Index: 930458 (C4 (Web Text))

Max Activating Token Index: 214

Text #2

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.6141. Min Act: -0.0000

Data Index: 978930 (C4 (Web Text))

Max Activating Token Index: 822

Text #3

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.6453. Min Act: -0.0000

Data Index: 170849 (C4 (Web Text))

Max Activating Token Index: 216

Text #4

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5716. Min Act: -0.0000

Data Index: 106030 (C4 (Web Text))

Max Activating Token Index: 642

Text #5

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5516. Min Act: -0.0000

Data Index: 690969 (C4 (Web Text))

Max Activating Token Index: 107

Text #6

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5530. Min Act: -0.0000

Data Index: 927515 (C4 (Web Text))

Max Activating Token Index: 180

Text #7

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5435. Min Act: -0.0000

Data Index: 241639 (C4 (Web Text))

Max Activating Token Index: 659

Text #8

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5359. Min Act: -0.0000

Data Index: 1264049 (C4 (Web Text))

Max Activating Token Index: 105

Text #9

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5488. Min Act: -0.0000

Data Index: 22162 (C4 (Web Text))

Max Activating Token Index: 722

Text #10

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5250. Min Act: -0.0000

Data Index: 1104315 (C4 (Web Text))

Max Activating Token Index: 119

Text #11

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5271. Min Act: -0.0000

Data Index: 1591735 (Python Code)

Max Activating Token Index: 712

Text #12

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.4888. Min Act: -0.0000

Data Index: 1313187 (C4 (Web Text))

Max Activating Token Index: 45

Text #13

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5009. Min Act: -0.0000

Data Index: 1021909 (C4 (Web Text))

Max Activating Token Index: 136

Text #14

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5435. Min Act: -0.0000

Data Index: 307068 (C4 (Web Text))

Max Activating Token Index: 781

Text #15

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5050. Min Act: -0.0000

Data Index: 1378453 (Python Code)

Max Activating Token Index: 124

Text #16

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5052. Min Act: -0.0000

Data Index: 1558968 (Python Code)

Max Activating Token Index: 456

Text #17

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.5216. Min Act: -0.0000

Data Index: 451291 (C4 (Web Text))

Max Activating Token Index: 842

Text #18

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.4848. Min Act: -0.0000

Data Index: 768369 (C4 (Web Text))

Max Activating Token Index: 534

Text #19

Max Range: 0.7624. Min Range: -0.7624

Max Act: 0.4824. Min Act: -0.0000

Data Index: 286121 (C4 (Web Text))

Max Activating Token Index: 139