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

Dataset: The Pile

Neuron 152 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: 7.3546. Min Range: -7.3546

Max Act: 6.9514. Min Act: -0.0000

Data Index: 626578 (The Pile)

Max Activating Token Index: 637

Text #1

Max Range: 7.3546. Min Range: -7.3546

Max Act: 7.2779. Min Act: -0.0000

Data Index: 1826874 (The Pile)

Max Activating Token Index: 802

Text #2

Max Range: 7.3546. Min Range: -7.3546

Max Act: 7.0715. Min Act: -0.0000

Data Index: 648635 (The Pile)

Max Activating Token Index: 488

Text #3

Max Range: 7.3546. Min Range: -7.3546

Max Act: 7.3546. Min Act: -0.0000

Data Index: 1241393 (The Pile)

Max Activating Token Index: 906

Text #4

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.7988. Min Act: -0.0000

Data Index: 621870 (The Pile)

Max Activating Token Index: 195

Text #5

Max Range: 7.3546. Min Range: -7.3546

Max Act: 7.2032. Min Act: -0.0000

Data Index: 286629 (The Pile)

Max Activating Token Index: 807

Text #6

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.6983. Min Act: -0.0000

Data Index: 903847 (The Pile)

Max Activating Token Index: 333

Text #7

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.6744. Min Act: -0.0000

Data Index: 1544672 (The Pile)

Max Activating Token Index: 990

Text #8

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.8324. Min Act: -0.0000

Data Index: 171818 (The Pile)

Max Activating Token Index: 941

Text #9

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.4279. Min Act: -0.0000

Data Index: 103850 (The Pile)

Max Activating Token Index: 284

Text #10

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.3242. Min Act: -0.0000

Data Index: 528338 (The Pile)

Max Activating Token Index: 256

Text #11

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.3177. Min Act: -0.0000

Data Index: 88276 (The Pile)

Max Activating Token Index: 107

Text #12

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.8419. Min Act: -0.0000

Data Index: 1954545 (The Pile)

Max Activating Token Index: 833

Text #13

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.3185. Min Act: -0.0000

Data Index: 450258 (The Pile)

Max Activating Token Index: 1008

Text #14

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.3991. Min Act: -0.0000

Data Index: 1469692 (The Pile)

Max Activating Token Index: 303

Text #15

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.5653. Min Act: -0.0000

Data Index: 364557 (The Pile)

Max Activating Token Index: 496

Text #16

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.1670. Min Act: -0.0000

Data Index: 1171311 (The Pile)

Max Activating Token Index: 592

Text #17

Max Range: 7.3546. Min Range: -7.3546

Max Act: 5.9419. Min Act: -0.0000

Data Index: 150163 (The Pile)

Max Activating Token Index: 596

Text #18

Max Range: 7.3546. Min Range: -7.3546

Max Act: 5.8987. Min Act: -0.0000

Data Index: 981343 (The Pile)

Max Activating Token Index: 675

Text #19

Max Range: 7.3546. Min Range: -7.3546

Max Act: 6.3143. Min Act: -0.0000

Data Index: 308493 (The Pile)

Max Activating Token Index: 1011