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

Dataset: The Pile

Neuron 602 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 9.0398. Min Range: -9.0398

Max Act: 9.0398. Min Act: -0.0000

Data Index: 155105 (The Pile)

Max Activating Token Index: 568

Text #1

Max Range: 9.0398. Min Range: -9.0398

Max Act: 8.4759. Min Act: -0.0000

Data Index: 1618891 (The Pile)

Max Activating Token Index: 780

Text #2

Max Range: 9.0398. Min Range: -9.0398

Max Act: 8.2675. Min Act: -0.0000

Data Index: 759878 (The Pile)

Max Activating Token Index: 214

Text #3

Max Range: 9.0398. Min Range: -9.0398

Max Act: 8.4557. Min Act: -0.0000

Data Index: 1062567 (The Pile)

Max Activating Token Index: 683

Text #4

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.9496. Min Act: -0.0000

Data Index: 42346 (The Pile)

Max Activating Token Index: 181

Text #5

Max Range: 9.0398. Min Range: -9.0398

Max Act: 8.1406. Min Act: 0.0000

Data Index: 1163902 (The Pile)

Max Activating Token Index: 517

Text #6

Max Range: 9.0398. Min Range: -9.0398

Max Act: 8.0076. Min Act: 0.0000

Data Index: 1901051 (The Pile)

Max Activating Token Index: 883

Text #7

Max Range: 9.0398. Min Range: -9.0398

Max Act: 8.1444. Min Act: 0.0001

Data Index: 1789149 (The Pile)

Max Activating Token Index: 445

Text #8

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.9525. Min Act: -0.0000

Data Index: 1454353 (The Pile)

Max Activating Token Index: 423

Text #9

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.7783. Min Act: -0.0000

Data Index: 856099 (The Pile)

Max Activating Token Index: 221

Text #10

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.6728. Min Act: -0.0000

Data Index: 1388100 (The Pile)

Max Activating Token Index: 337

Text #11

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.8372. Min Act: -0.0000

Data Index: 200505 (The Pile)

Max Activating Token Index: 271

Text #12

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.9005. Min Act: -0.0000

Data Index: 629361 (The Pile)

Max Activating Token Index: 564

Text #13

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.8820. Min Act: -0.0000

Data Index: 1494093 (The Pile)

Max Activating Token Index: 551

Text #14

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.4471. Min Act: -0.0000

Data Index: 1403449 (The Pile)

Max Activating Token Index: 438

Text #15

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.7887. Min Act: -0.0000

Data Index: 1532345 (The Pile)

Max Activating Token Index: 349

Text #16

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.6049. Min Act: -0.0000

Data Index: 1382911 (The Pile)

Max Activating Token Index: 158

Text #17

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.7878. Min Act: 0.0001

Data Index: 420070 (The Pile)

Max Activating Token Index: 565

Text #18

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.5174. Min Act: 0.0000

Data Index: 1072836 (The Pile)

Max Activating Token Index: 505

Text #19

Max Range: 9.0398. Min Range: -9.0398

Max Act: 7.4691. Min Act: -0.0000

Data Index: 1658834 (The Pile)

Max Activating Token Index: 999