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

Dataset: The Pile

Neuron 144 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: 10.2736. Min Range: -10.2736

Max Act: 10.2736. Min Act: -0.0000

Data Index: 1432358 (The Pile)

Max Activating Token Index: 524

Text #1

Max Range: 10.2736. Min Range: -10.2736

Max Act: 10.0717. Min Act: -0.0000

Data Index: 452399 (The Pile)

Max Activating Token Index: 471

Text #2

Max Range: 10.2736. Min Range: -10.2736

Max Act: 9.8357. Min Act: -0.0000

Data Index: 629218 (The Pile)

Max Activating Token Index: 337

Text #3

Max Range: 10.2736. Min Range: -10.2736

Max Act: 9.7862. Min Act: -0.0000

Data Index: 1627896 (The Pile)

Max Activating Token Index: 458

Text #4

Max Range: 10.2736. Min Range: -10.2736

Max Act: 9.6930. Min Act: -0.0000

Data Index: 496339 (The Pile)

Max Activating Token Index: 464

Text #5

Max Range: 10.2736. Min Range: -10.2736

Max Act: 9.4000. Min Act: -0.0000

Data Index: 1274510 (The Pile)

Max Activating Token Index: 560

Text #6

Max Range: 10.2736. Min Range: -10.2736

Max Act: 9.0725. Min Act: -0.0000

Data Index: 1658381 (The Pile)

Max Activating Token Index: 501

Text #7

Max Range: 10.2736. Min Range: -10.2736

Max Act: 8.9908. Min Act: -0.0000

Data Index: 910460 (The Pile)

Max Activating Token Index: 357

Text #8

Max Range: 10.2736. Min Range: -10.2736

Max Act: 8.4820. Min Act: -0.0000

Data Index: 1618133 (The Pile)

Max Activating Token Index: 479

Text #9

Max Range: 10.2736. Min Range: -10.2736

Max Act: 8.3937. Min Act: -0.0000

Data Index: 237542 (The Pile)

Max Activating Token Index: 411

Text #10

Max Range: 10.2736. Min Range: -10.2736

Max Act: 8.7089. Min Act: -0.0000

Data Index: 1453773 (The Pile)

Max Activating Token Index: 884

Text #11

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.8825. Min Act: -0.0000

Data Index: 1924471 (The Pile)

Max Activating Token Index: 851

Text #12

Max Range: 10.2736. Min Range: -10.2736

Max Act: 8.4674. Min Act: -0.0000

Data Index: 1456217 (The Pile)

Max Activating Token Index: 667

Text #13

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.6559. Min Act: -0.0000

Data Index: 966506 (The Pile)

Max Activating Token Index: 104

Text #14

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.1163. Min Act: -0.0000

Data Index: 1082132 (The Pile)

Max Activating Token Index: 352

Text #15

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.7432. Min Act: -0.0000

Data Index: 433246 (The Pile)

Max Activating Token Index: 876

Text #16

Max Range: 10.2736. Min Range: -10.2736

Max Act: 6.9368. Min Act: -0.0000

Data Index: 919353 (The Pile)

Max Activating Token Index: 74

Text #17

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.2608. Min Act: -0.0000

Data Index: 811813 (The Pile)

Max Activating Token Index: 180

Text #18

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.5948. Min Act: -0.0000

Data Index: 1267164 (The Pile)

Max Activating Token Index: 767

Text #19

Max Range: 10.2736. Min Range: -10.2736

Max Act: 7.3751. Min Act: -0.0000

Data Index: 306416 (The Pile)

Max Activating Token Index: 386