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

Dataset: The Pile

Neuron 3673 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.6804. Min Act: -0.0000

Data Index: 426627 (The Pile)

Max Activating Token Index: 136

Text #1

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.7076. Min Act: -0.0000

Data Index: 448212 (The Pile)

Max Activating Token Index: 295

Text #2

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.6174. Min Act: -0.0000

Data Index: 774634 (The Pile)

Max Activating Token Index: 16

Text #3

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5827. Min Act: -0.0000

Data Index: 1536521 (The Pile)

Max Activating Token Index: 16

Text #4

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.6078. Min Act: -0.0000

Data Index: 763452 (The Pile)

Max Activating Token Index: 265

Text #5

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5684. Min Act: -0.0000

Data Index: 1986597 (The Pile)

Max Activating Token Index: 455

Text #6

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5249. Min Act: -0.0000

Data Index: 498854 (The Pile)

Max Activating Token Index: 305

Text #7

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5222. Min Act: -0.0000

Data Index: 404120 (The Pile)

Max Activating Token Index: 322

Text #8

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5393. Min Act: -0.0000

Data Index: 1799470 (The Pile)

Max Activating Token Index: 189

Text #9

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5797. Min Act: -0.0000

Data Index: 27353 (The Pile)

Max Activating Token Index: 670

Text #10

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5577. Min Act: -0.0000

Data Index: 725029 (The Pile)

Max Activating Token Index: 161

Text #11

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5355. Min Act: -0.0000

Data Index: 1581764 (The Pile)

Max Activating Token Index: 692

Text #12

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5321. Min Act: -0.0000

Data Index: 320664 (The Pile)

Max Activating Token Index: 621

Text #13

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5618. Min Act: -0.0000

Data Index: 36792 (The Pile)

Max Activating Token Index: 986

Text #14

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5103. Min Act: -0.0000

Data Index: 446346 (The Pile)

Max Activating Token Index: 286

Text #15

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5424. Min Act: -0.0000

Data Index: 1561284 (The Pile)

Max Activating Token Index: 939

Text #16

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5358. Min Act: -0.0000

Data Index: 361843 (The Pile)

Max Activating Token Index: 678

Text #17

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5019. Min Act: -0.0000

Data Index: 67380 (The Pile)

Max Activating Token Index: 278

Text #18

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5140. Min Act: -0.0000

Data Index: 1324327 (The Pile)

Max Activating Token Index: 303

Text #19

Max Range: 0.7076. Min Range: -0.7076

Max Act: 0.5127. Min Act: -0.0000

Data Index: 1026976 (The Pile)

Max Activating Token Index: 199