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

Dataset: The Pile

Neuron 1607 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: 10.3349. Min Range: -10.3349

Max Act: 10.3349. Min Act: -0.0000

Data Index: 1884072 (The Pile)

Max Activating Token Index: 863

Text #1

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.7613. Min Act: -0.0000

Data Index: 1331020 (The Pile)

Max Activating Token Index: 305

Text #2

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.8953. Min Act: -0.0000

Data Index: 1572842 (The Pile)

Max Activating Token Index: 413

Text #3

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.5656. Min Act: -0.0000

Data Index: 558277 (The Pile)

Max Activating Token Index: 472

Text #4

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.5273. Min Act: -0.0000

Data Index: 335869 (The Pile)

Max Activating Token Index: 309

Text #5

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.3262. Min Act: -0.0000

Data Index: 1381363 (The Pile)

Max Activating Token Index: 433

Text #6

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.3396. Min Act: -0.0000

Data Index: 1842854 (The Pile)

Max Activating Token Index: 91

Text #7

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.5568. Min Act: -0.0000

Data Index: 1839335 (The Pile)

Max Activating Token Index: 488

Text #8

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.1448. Min Act: -0.0000

Data Index: 52838 (The Pile)

Max Activating Token Index: 633

Text #9

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.1477. Min Act: -0.0000

Data Index: 1092625 (The Pile)

Max Activating Token Index: 527

Text #10

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.7203. Min Act: -0.0000

Data Index: 594976 (The Pile)

Max Activating Token Index: 962

Text #11

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.9126. Min Act: -0.0000

Data Index: 1842084 (The Pile)

Max Activating Token Index: 784

Text #12

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.8892. Min Act: -0.0000

Data Index: 518909 (The Pile)

Max Activating Token Index: 333

Text #13

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.6679. Min Act: -0.0000

Data Index: 374791 (The Pile)

Max Activating Token Index: 790

Text #14

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.8158. Min Act: -0.0000

Data Index: 1090605 (The Pile)

Max Activating Token Index: 660

Text #15

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.7602. Min Act: -0.0000

Data Index: 534804 (The Pile)

Max Activating Token Index: 623

Text #16

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.7243. Min Act: -0.0000

Data Index: 1229485 (The Pile)

Max Activating Token Index: 540

Text #17

Max Range: 10.3349. Min Range: -10.3349

Max Act: 9.0819. Min Act: -0.0000

Data Index: 507672 (The Pile)

Max Activating Token Index: 1012

Text #18

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.5641. Min Act: -0.0000

Data Index: 336589 (The Pile)

Max Activating Token Index: 243

Text #19

Max Range: 10.3349. Min Range: -10.3349

Max Act: 8.8232. Min Act: -0.0000

Data Index: 1487803 (The Pile)

Max Activating Token Index: 356