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

Dataset: The Pile

Neuron 205 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.8611. Min Act: -0.0000

Data Index: 1619702 (The Pile)

Max Activating Token Index: 779

Text #1

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.3357. Min Act: -0.0000

Data Index: 348042 (The Pile)

Max Activating Token Index: 764

Text #2

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.5362. Min Act: -0.0000

Data Index: 1870383 (The Pile)

Max Activating Token Index: 268

Text #3

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.5086. Min Act: -0.0000

Data Index: 1647054 (The Pile)

Max Activating Token Index: 908

Text #4

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.2473. Min Act: -0.0000

Data Index: 424587 (The Pile)

Max Activating Token Index: 812

Text #5

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.5483. Min Act: -0.0000

Data Index: 1068260 (The Pile)

Max Activating Token Index: 488

Text #6

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.2963. Min Act: -0.0000

Data Index: 1191479 (The Pile)

Max Activating Token Index: 612

Text #7

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.1778. Min Act: -0.0000

Data Index: 1746270 (The Pile)

Max Activating Token Index: 979

Text #8

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.2610. Min Act: -0.0000

Data Index: 553002 (The Pile)

Max Activating Token Index: 584

Text #9

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.2588. Min Act: -0.0000

Data Index: 1342635 (The Pile)

Max Activating Token Index: 284

Text #10

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.2401. Min Act: -0.0000

Data Index: 1339047 (The Pile)

Max Activating Token Index: 298

Text #11

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.1318. Min Act: -0.0000

Data Index: 1847783 (The Pile)

Max Activating Token Index: 12

Text #12

Max Range: 11.8611. Min Range: -11.8611

Max Act: 10.9836. Min Act: -0.0000

Data Index: 1668349 (The Pile)

Max Activating Token Index: 324

Text #13

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.1493. Min Act: -0.0000

Data Index: 1397792 (The Pile)

Max Activating Token Index: 415

Text #14

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.1358. Min Act: -0.0000

Data Index: 1210268 (The Pile)

Max Activating Token Index: 466

Text #15

Max Range: 11.8611. Min Range: -11.8611

Max Act: 10.9927. Min Act: -0.0000

Data Index: 1382833 (The Pile)

Max Activating Token Index: 62

Text #16

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.0393. Min Act: -0.0000

Data Index: 860549 (The Pile)

Max Activating Token Index: 49

Text #17

Max Range: 11.8611. Min Range: -11.8611

Max Act: 10.7321. Min Act: -0.0000

Data Index: 1691993 (The Pile)

Max Activating Token Index: 877

Text #18

Max Range: 11.8611. Min Range: -11.8611

Max Act: 10.8560. Min Act: 0.0001

Data Index: 832863 (The Pile)

Max Activating Token Index: 169

Text #19

Max Range: 11.8611. Min Range: -11.8611

Max Act: 11.0868. Min Act: -0.0000

Data Index: 243833 (The Pile)

Max Activating Token Index: 109