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

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

Neuron 2476 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.2508. Min Act: -0.0000

Data Index: 665274 (C4 (Web Text))

Max Activating Token Index: 54

Text #1

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.2306. Min Act: -0.0000

Data Index: 10320 (C4 (Web Text))

Max Activating Token Index: 283

Text #2

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1870. Min Act: -0.0000

Data Index: 446760 (C4 (Web Text))

Max Activating Token Index: 136

Text #3

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1798. Min Act: -0.0000

Data Index: 1304930 (C4 (Web Text))

Max Activating Token Index: 841

Text #4

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1847. Min Act: -0.0000

Data Index: 206482 (C4 (Web Text))

Max Activating Token Index: 920

Text #5

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1764. Min Act: -0.0000

Data Index: 525936 (C4 (Web Text))

Max Activating Token Index: 917

Text #6

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1741. Min Act: -0.0000

Data Index: 510640 (C4 (Web Text))

Max Activating Token Index: 363

Text #7

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1907. Min Act: -0.0000

Data Index: 821684 (C4 (Web Text))

Max Activating Token Index: 521

Text #8

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1778. Min Act: -0.0000

Data Index: 395154 (C4 (Web Text))

Max Activating Token Index: 782

Text #9

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1928. Min Act: -0.0000

Data Index: 1280275 (C4 (Web Text))

Max Activating Token Index: 276

Text #10

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1970. Min Act: -0.0000

Data Index: 844162 (C4 (Web Text))

Max Activating Token Index: 971

Text #11

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1679. Min Act: -0.0000

Data Index: 1189004 (C4 (Web Text))

Max Activating Token Index: 553

Text #12

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1734. Min Act: -0.0000

Data Index: 550660 (C4 (Web Text))

Max Activating Token Index: 969

Text #13

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1832. Min Act: -0.0001

Data Index: 1633327 (Python Code)

Max Activating Token Index: 238

Text #14

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1578. Min Act: -0.0000

Data Index: 1311188 (C4 (Web Text))

Max Activating Token Index: 380

Text #15

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1684. Min Act: -0.0000

Data Index: 372602 (C4 (Web Text))

Max Activating Token Index: 150

Text #16

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1678. Min Act: -0.0000

Data Index: 553737 (C4 (Web Text))

Max Activating Token Index: 521

Text #17

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1550. Min Act: -0.0000

Data Index: 1263415 (C4 (Web Text))

Max Activating Token Index: 277

Text #18

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1662. Min Act: -0.0000

Data Index: 930617 (C4 (Web Text))

Max Activating Token Index: 939

Text #19

Max Range: 0.2508. Min Range: -0.2508

Max Act: 0.1606. Min Act: -0.0000

Data Index: 1202536 (C4 (Web Text))

Max Activating Token Index: 773