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

Dataset: The Pile

Neuron 665 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: 5.4853. Min Range: -5.4853

Max Act: 5.2841. Min Act: -0.0000

Data Index: 722913 (The Pile)

Max Activating Token Index: 242

Text #1

Max Range: 5.4853. Min Range: -5.4853

Max Act: 5.4853. Min Act: -0.0000

Data Index: 133086 (The Pile)

Max Activating Token Index: 594

Text #2

Max Range: 5.4853. Min Range: -5.4853

Max Act: 5.2551. Min Act: -0.0000

Data Index: 1420934 (The Pile)

Max Activating Token Index: 418

Text #3

Max Range: 5.4853. Min Range: -5.4853

Max Act: 5.1231. Min Act: -0.0000

Data Index: 425455 (The Pile)

Max Activating Token Index: 247

Text #4

Max Range: 5.4853. Min Range: -5.4853

Max Act: 5.3525. Min Act: -0.0000

Data Index: 554131 (The Pile)

Max Activating Token Index: 32

Text #5

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.9163. Min Act: -0.0000

Data Index: 1384106 (The Pile)

Max Activating Token Index: 376

Text #6

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.8596. Min Act: -0.0000

Data Index: 133069 (The Pile)

Max Activating Token Index: 500

Text #7

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.9495. Min Act: -0.0000

Data Index: 896403 (The Pile)

Max Activating Token Index: 515

Text #8

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.6351. Min Act: -0.0000

Data Index: 1283764 (The Pile)

Max Activating Token Index: 372

Text #9

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.7822. Min Act: -0.0000

Data Index: 1615201 (The Pile)

Max Activating Token Index: 522

Text #10

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.9118. Min Act: -0.0000

Data Index: 1950233 (The Pile)

Max Activating Token Index: 205

Text #11

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.6786. Min Act: -0.0000

Data Index: 291098 (The Pile)

Max Activating Token Index: 108

Text #12

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.8243. Min Act: -0.0000

Data Index: 299900 (The Pile)

Max Activating Token Index: 39

Text #13

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.4820. Min Act: -0.0000

Data Index: 1538952 (The Pile)

Max Activating Token Index: 273

Text #14

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.9913. Min Act: -0.0000

Data Index: 638369 (The Pile)

Max Activating Token Index: 321

Text #15

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.6478. Min Act: -0.0000

Data Index: 478756 (The Pile)

Max Activating Token Index: 406

Text #16

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.2500. Min Act: -0.0000

Data Index: 280493 (The Pile)

Max Activating Token Index: 637

Text #17

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.6130. Min Act: -0.0000

Data Index: 1720372 (The Pile)

Max Activating Token Index: 82

Text #18

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.6094. Min Act: -0.0000

Data Index: 1150810 (The Pile)

Max Activating Token Index: 224

Text #19

Max Range: 5.4853. Min Range: -5.4853

Max Act: 4.5866. Min Act: -0.0000

Data Index: 576105 (The Pile)

Max Activating Token Index: 449