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

Dataset: The Pile

Neuron 3322 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: 7.2078. Min Range: -7.2078

Max Act: 7.1732. Min Act: -0.0000

Data Index: 1809374 (The Pile)

Max Activating Token Index: 44

Text #1

Max Range: 7.2078. Min Range: -7.2078

Max Act: 7.0396. Min Act: -0.0000

Data Index: 1638266 (The Pile)

Max Activating Token Index: 376

Text #2

Max Range: 7.2078. Min Range: -7.2078

Max Act: 7.2078. Min Act: -0.0000

Data Index: 1286386 (The Pile)

Max Activating Token Index: 60

Text #3

Max Range: 7.2078. Min Range: -7.2078

Max Act: 7.0551. Min Act: -0.0000

Data Index: 82392 (The Pile)

Max Activating Token Index: 704

Text #4

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.2238. Min Act: -0.0000

Data Index: 533692 (The Pile)

Max Activating Token Index: 66

Text #5

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.8543. Min Act: -0.0000

Data Index: 868128 (The Pile)

Max Activating Token Index: 51

Text #6

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.4466. Min Act: -0.0000

Data Index: 221457 (The Pile)

Max Activating Token Index: 654

Text #7

Max Range: 7.2078. Min Range: -7.2078

Max Act: 7.1832. Min Act: -0.0000

Data Index: 944100 (The Pile)

Max Activating Token Index: 590

Text #8

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.6834. Min Act: -0.0000

Data Index: 784107 (The Pile)

Max Activating Token Index: 853

Text #9

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.2142. Min Act: -0.0000

Data Index: 1819953 (The Pile)

Max Activating Token Index: 969

Text #10

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.3913. Min Act: -0.0000

Data Index: 321066 (The Pile)

Max Activating Token Index: 648

Text #11

Max Range: 7.2078. Min Range: -7.2078

Max Act: 5.9821. Min Act: -0.0000

Data Index: 543712 (The Pile)

Max Activating Token Index: 857

Text #12

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.9722. Min Act: -0.0000

Data Index: 1816280 (The Pile)

Max Activating Token Index: 818

Text #13

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.4928. Min Act: -0.0000

Data Index: 891934 (The Pile)

Max Activating Token Index: 860

Text #14

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.7031. Min Act: -0.0000

Data Index: 530564 (The Pile)

Max Activating Token Index: 129

Text #15

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.4494. Min Act: -0.0000

Data Index: 1549540 (The Pile)

Max Activating Token Index: 60

Text #16

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.2571. Min Act: -0.0000

Data Index: 1219867 (The Pile)

Max Activating Token Index: 51

Text #17

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.1988. Min Act: -0.0000

Data Index: 1656205 (The Pile)

Max Activating Token Index: 30

Text #18

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.0198. Min Act: -0.0000

Data Index: 922926 (The Pile)

Max Activating Token Index: 538

Text #19

Max Range: 7.2078. Min Range: -7.2078

Max Act: 6.2445. Min Act: -0.0000

Data Index: 1546180 (The Pile)

Max Activating Token Index: 298