Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 2720 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.9162. Min Act: -0.0000

Data Index: 1785793 (The Pile)

Max Activating Token Index: 98

Text #1

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.8220. Min Act: -0.0000

Data Index: 520145 (The Pile)

Max Activating Token Index: 167

Text #2

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6675. Min Act: -0.0000

Data Index: 1720425 (The Pile)

Max Activating Token Index: 55

Text #3

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6851. Min Act: -0.0000

Data Index: 1125077 (The Pile)

Max Activating Token Index: 332

Text #4

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6805. Min Act: -0.0000

Data Index: 372889 (The Pile)

Max Activating Token Index: 443

Text #5

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6240. Min Act: -0.0000

Data Index: 1858851 (The Pile)

Max Activating Token Index: 638

Text #6

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6839. Min Act: -0.0000

Data Index: 1632902 (The Pile)

Max Activating Token Index: 58

Text #7

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6876. Min Act: -0.0000

Data Index: 285805 (The Pile)

Max Activating Token Index: 173

Text #8

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6840. Min Act: -0.0000

Data Index: 1881138 (The Pile)

Max Activating Token Index: 353

Text #9

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6384. Min Act: -0.0000

Data Index: 1420147 (The Pile)

Max Activating Token Index: 432

Text #10

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6790. Min Act: -0.0000

Data Index: 480362 (The Pile)

Max Activating Token Index: 245

Text #11

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6045. Min Act: -0.0000

Data Index: 889141 (The Pile)

Max Activating Token Index: 876

Text #12

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6580. Min Act: -0.0000

Data Index: 247112 (The Pile)

Max Activating Token Index: 541

Text #13

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6973. Min Act: -0.0000

Data Index: 106611 (The Pile)

Max Activating Token Index: 308

Text #14

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.5978. Min Act: -0.0000

Data Index: 438076 (The Pile)

Max Activating Token Index: 28

Text #15

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6497. Min Act: -0.0000

Data Index: 507357 (The Pile)

Max Activating Token Index: 392

Text #16

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6256. Min Act: -0.0000

Data Index: 578727 (The Pile)

Max Activating Token Index: 130

Text #17

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6166. Min Act: -0.0000

Data Index: 1157659 (The Pile)

Max Activating Token Index: 56

Text #18

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6300. Min Act: -0.0000

Data Index: 43717 (The Pile)

Max Activating Token Index: 28

Text #19

Max Range: 0.9162. Min Range: -0.9162

Max Act: 0.6293. Min Act: -0.0000

Data Index: 1821405 (The Pile)

Max Activating Token Index: 309