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

Dataset: The Pile

Neuron 2893 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.1833. Min Range: -8.1833

Max Act: 8.1833. Min Act: -0.0000

Data Index: 815996 (The Pile)

Max Activating Token Index: 89

Text #1

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.9045. Min Act: -0.0000

Data Index: 392616 (The Pile)

Max Activating Token Index: 440

Text #2

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.5274. Min Act: -0.0000

Data Index: 1255084 (The Pile)

Max Activating Token Index: 243

Text #3

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.6827. Min Act: -0.0000

Data Index: 207915 (The Pile)

Max Activating Token Index: 449

Text #4

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.4882. Min Act: -0.0000

Data Index: 1748890 (The Pile)

Max Activating Token Index: 362

Text #5

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.4116. Min Act: -0.0000

Data Index: 91173 (The Pile)

Max Activating Token Index: 467

Text #6

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.3794. Min Act: -0.0000

Data Index: 1995908 (The Pile)

Max Activating Token Index: 234

Text #7

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.5202. Min Act: -0.0000

Data Index: 364842 (The Pile)

Max Activating Token Index: 190

Text #8

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.1361. Min Act: -0.0000

Data Index: 1640622 (The Pile)

Max Activating Token Index: 189

Text #9

Max Range: 8.1833. Min Range: -8.1833

Max Act: 7.0569. Min Act: -0.0000

Data Index: 260457 (The Pile)

Max Activating Token Index: 267

Text #10

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.9622. Min Act: -0.0000

Data Index: 640529 (The Pile)

Max Activating Token Index: 724

Text #11

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.9150. Min Act: -0.0000

Data Index: 1383915 (The Pile)

Max Activating Token Index: 488

Text #12

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.8354. Min Act: -0.0000

Data Index: 1067451 (The Pile)

Max Activating Token Index: 163

Text #13

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.8217. Min Act: -0.0000

Data Index: 832639 (The Pile)

Max Activating Token Index: 427

Text #14

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.8346. Min Act: -0.0000

Data Index: 374761 (The Pile)

Max Activating Token Index: 575

Text #15

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.8115. Min Act: -0.0000

Data Index: 1441824 (The Pile)

Max Activating Token Index: 273

Text #16

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.9642. Min Act: -0.0000

Data Index: 442596 (The Pile)

Max Activating Token Index: 359

Text #17

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.6375. Min Act: -0.0000

Data Index: 763171 (The Pile)

Max Activating Token Index: 301

Text #18

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.4041. Min Act: -0.0000

Data Index: 385407 (The Pile)

Max Activating Token Index: 645

Text #19

Max Range: 8.1833. Min Range: -8.1833

Max Act: 6.5580. Min Act: -0.0000

Data Index: 1149511 (The Pile)

Max Activating Token Index: 535