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

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

Neuron 5171 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: 0.7100. Min Range: -0.7100

Max Act: 0.7100. Min Act: -0.0000

Data Index: 404792 (C4 (Web Text))

Max Activating Token Index: 852

Text #1

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.6721. Min Act: -0.0000

Data Index: 1487542 (Python Code)

Max Activating Token Index: 100

Text #2

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.6882. Min Act: -0.0000

Data Index: 954768 (C4 (Web Text))

Max Activating Token Index: 851

Text #3

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.6395. Min Act: -0.0000

Data Index: 1119895 (C4 (Web Text))

Max Activating Token Index: 426

Text #4

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.7042. Min Act: -0.0000

Data Index: 1470846 (Python Code)

Max Activating Token Index: 275

Text #5

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.6115. Min Act: -0.0000

Data Index: 627200 (C4 (Web Text))

Max Activating Token Index: 558

Text #6

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.6152. Min Act: 0.0000

Data Index: 1499340 (Python Code)

Max Activating Token Index: 35

Text #7

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.6203. Min Act: -0.0000

Data Index: 969315 (C4 (Web Text))

Max Activating Token Index: 145

Text #8

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5757. Min Act: -0.0000

Data Index: 953047 (C4 (Web Text))

Max Activating Token Index: 860

Text #9

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5979. Min Act: -0.0000

Data Index: 1316009 (C4 (Web Text))

Max Activating Token Index: 792

Text #10

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5640. Min Act: -0.0000

Data Index: 158690 (C4 (Web Text))

Max Activating Token Index: 190

Text #11

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5171. Min Act: -0.0000

Data Index: 755593 (C4 (Web Text))

Max Activating Token Index: 895

Text #12

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5582. Min Act: -0.0000

Data Index: 1171507 (C4 (Web Text))

Max Activating Token Index: 684

Text #13

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5720. Min Act: -0.0000

Data Index: 1462711 (Python Code)

Max Activating Token Index: 611

Text #14

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5479. Min Act: -0.0000

Data Index: 999609 (C4 (Web Text))

Max Activating Token Index: 23

Text #15

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5376. Min Act: -0.0000

Data Index: 1531024 (Python Code)

Max Activating Token Index: 244

Text #16

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5114. Min Act: -0.0000

Data Index: 1436306 (Python Code)

Max Activating Token Index: 603

Text #17

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.5311. Min Act: -0.0000

Data Index: 1047065 (C4 (Web Text))

Max Activating Token Index: 368

Text #18

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.4966. Min Act: -0.0000

Data Index: 705970 (C4 (Web Text))

Max Activating Token Index: 755

Text #19

Max Range: 0.7100. Min Range: -0.7100

Max Act: 0.4885. Min Act: -0.0000

Data Index: 82944 (C4 (Web Text))

Max Activating Token Index: 903