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

Dataset: The Pile

Neuron 1008 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: 8.5178. Min Range: -8.5178

Max Act: 8.5178. Min Act: -0.0000

Data Index: 1278811 (The Pile)

Max Activating Token Index: 167

Text #1

Max Range: 8.5178. Min Range: -8.5178

Max Act: 8.3093. Min Act: -0.0000

Data Index: 1695814 (The Pile)

Max Activating Token Index: 147

Text #2

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.7818. Min Act: -0.0000

Data Index: 1013404 (The Pile)

Max Activating Token Index: 270

Text #3

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.9708. Min Act: -0.0000

Data Index: 335044 (The Pile)

Max Activating Token Index: 792

Text #4

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.8759. Min Act: -0.0000

Data Index: 622773 (The Pile)

Max Activating Token Index: 227

Text #5

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.6244. Min Act: -0.0000

Data Index: 1527739 (The Pile)

Max Activating Token Index: 522

Text #6

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.7097. Min Act: -0.0000

Data Index: 731121 (The Pile)

Max Activating Token Index: 387

Text #7

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.3846. Min Act: 0.0000

Data Index: 1933298 (The Pile)

Max Activating Token Index: 976

Text #8

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.3347. Min Act: -0.0000

Data Index: 1842237 (The Pile)

Max Activating Token Index: 153

Text #9

Max Range: 8.5178. Min Range: -8.5178

Max Act: 6.6056. Min Act: 0.0000

Data Index: 1681722 (The Pile)

Max Activating Token Index: 790

Text #10

Max Range: 8.5178. Min Range: -8.5178

Max Act: 7.1251. Min Act: -0.0000

Data Index: 1672188 (The Pile)

Max Activating Token Index: 821

Text #11

Max Range: 8.5178. Min Range: -8.5178

Max Act: 6.7681. Min Act: -0.0000

Data Index: 1848531 (The Pile)

Max Activating Token Index: 87

Text #12

Max Range: 8.5178. Min Range: -8.5178

Max Act: 6.5115. Min Act: -0.0000

Data Index: 1466529 (The Pile)

Max Activating Token Index: 563

Text #13

Max Range: 8.5178. Min Range: -8.5178

Max Act: 5.4048. Min Act: -0.0000

Data Index: 788904 (The Pile)

Max Activating Token Index: 124

Text #14

Max Range: 8.5178. Min Range: -8.5178

Max Act: 4.9241. Min Act: -0.0000

Data Index: 732168 (The Pile)

Max Activating Token Index: 285

Text #15

Max Range: 8.5178. Min Range: -8.5178

Max Act: 4.9201. Min Act: -0.0000

Data Index: 418261 (The Pile)

Max Activating Token Index: 110

Text #16

Max Range: 8.5178. Min Range: -8.5178

Max Act: 5.2178. Min Act: -0.0000

Data Index: 621673 (The Pile)

Max Activating Token Index: 1019

Text #17

Max Range: 8.5178. Min Range: -8.5178

Max Act: 4.7311. Min Act: -0.0000

Data Index: 199595 (The Pile)

Max Activating Token Index: 985

Text #18

Max Range: 8.5178. Min Range: -8.5178

Max Act: 4.5268. Min Act: 0.0000

Data Index: 1031964 (The Pile)

Max Activating Token Index: 488

Text #19

Max Range: 8.5178. Min Range: -8.5178

Max Act: 4.7527. Min Act: 0.0000

Data Index: 1944540 (The Pile)

Max Activating Token Index: 119