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

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

Neuron 1242 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: 0.2352. Min Range: -0.2352

Max Act: 0.2251. Min Act: -0.0001

Data Index: 1363153 (Python Code)

Max Activating Token Index: 119

Text #1

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.2352. Min Act: -0.0001

Data Index: 1077615 (C4 (Web Text))

Max Activating Token Index: 349

Text #2

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.2035. Min Act: -0.0001

Data Index: 161048 (C4 (Web Text))

Max Activating Token Index: 764

Text #3

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1835. Min Act: -0.0001

Data Index: 307552 (C4 (Web Text))

Max Activating Token Index: 222

Text #4

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1719. Min Act: -0.0001

Data Index: 1618459 (Python Code)

Max Activating Token Index: 853

Text #5

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1806. Min Act: -0.0001

Data Index: 1522751 (Python Code)

Max Activating Token Index: 805

Text #6

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1840. Min Act: -0.0001

Data Index: 143486 (C4 (Web Text))

Max Activating Token Index: 718

Text #7

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1864. Min Act: -0.0001

Data Index: 894853 (C4 (Web Text))

Max Activating Token Index: 425

Text #8

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1767. Min Act: -0.0001

Data Index: 964483 (C4 (Web Text))

Max Activating Token Index: 438

Text #9

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1706. Min Act: -0.0001

Data Index: 1255186 (C4 (Web Text))

Max Activating Token Index: 16

Text #10

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1713. Min Act: -0.0001

Data Index: 921371 (C4 (Web Text))

Max Activating Token Index: 622

Text #11

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1762. Min Act: -0.0001

Data Index: 1568741 (Python Code)

Max Activating Token Index: 72

Text #12

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1692. Min Act: -0.0000

Data Index: 720541 (C4 (Web Text))

Max Activating Token Index: 909

Text #13

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1630. Min Act: -0.0001

Data Index: 964423 (C4 (Web Text))

Max Activating Token Index: 735

Text #14

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1722. Min Act: -0.0001

Data Index: 1444208 (Python Code)

Max Activating Token Index: 217

Text #15

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1691. Min Act: -0.0001

Data Index: 560516 (C4 (Web Text))

Max Activating Token Index: 167

Text #16

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1637. Min Act: -0.0001

Data Index: 949505 (C4 (Web Text))

Max Activating Token Index: 313

Text #17

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1640. Min Act: -0.0001

Data Index: 1223572 (C4 (Web Text))

Max Activating Token Index: 190

Text #18

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1449. Min Act: -0.0001

Data Index: 60949 (C4 (Web Text))

Max Activating Token Index: 941

Text #19

Max Range: 0.2352. Min Range: -0.2352

Max Act: 0.1439. Min Act: -0.0001

Data Index: 1113858 (C4 (Web Text))

Max Activating Token Index: 712