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

Dataset: The Pile

Neuron 1842 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: 8.7104. Min Range: -8.7104

Max Act: 8.7104. Min Act: -0.0000

Data Index: 973599 (The Pile)

Max Activating Token Index: 755

Text #1

Max Range: 8.7104. Min Range: -8.7104

Max Act: 7.3486. Min Act: -0.0000

Data Index: 134701 (The Pile)

Max Activating Token Index: 698

Text #2

Max Range: 8.7104. Min Range: -8.7104

Max Act: 7.3238. Min Act: -0.0000

Data Index: 666380 (The Pile)

Max Activating Token Index: 750

Text #3

Max Range: 8.7104. Min Range: -8.7104

Max Act: 7.3956. Min Act: -0.0000

Data Index: 1261225 (The Pile)

Max Activating Token Index: 219

Text #4

Max Range: 8.7104. Min Range: -8.7104

Max Act: 7.3451. Min Act: -0.0000

Data Index: 1635052 (The Pile)

Max Activating Token Index: 728

Text #5

Max Range: 8.7104. Min Range: -8.7104

Max Act: 7.1445. Min Act: -0.0000

Data Index: 1249009 (The Pile)

Max Activating Token Index: 641

Text #6

Max Range: 8.7104. Min Range: -8.7104

Max Act: 7.3416. Min Act: -0.0000

Data Index: 1625367 (The Pile)

Max Activating Token Index: 337

Text #7

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.8599. Min Act: -0.0000

Data Index: 1464054 (The Pile)

Max Activating Token Index: 553

Text #8

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.3458. Min Act: -0.0000

Data Index: 120079 (The Pile)

Max Activating Token Index: 27

Text #9

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.7144. Min Act: -0.0000

Data Index: 1083714 (The Pile)

Max Activating Token Index: 704

Text #10

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.4222. Min Act: -0.0000

Data Index: 606069 (The Pile)

Max Activating Token Index: 445

Text #11

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.0275. Min Act: -0.0000

Data Index: 832127 (The Pile)

Max Activating Token Index: 730

Text #12

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.3682. Min Act: -0.0000

Data Index: 1114834 (The Pile)

Max Activating Token Index: 606

Text #13

Max Range: 8.7104. Min Range: -8.7104

Max Act: 6.2267. Min Act: -0.0000

Data Index: 962497 (The Pile)

Max Activating Token Index: 329

Text #14

Max Range: 8.7104. Min Range: -8.7104

Max Act: 5.4247. Min Act: -0.0000

Data Index: 1865064 (The Pile)

Max Activating Token Index: 180

Text #15

Max Range: 8.7104. Min Range: -8.7104

Max Act: 5.4709. Min Act: -0.0000

Data Index: 1872956 (The Pile)

Max Activating Token Index: 471

Text #16

Max Range: 8.7104. Min Range: -8.7104

Max Act: 5.5343. Min Act: -0.0000

Data Index: 1666438 (The Pile)

Max Activating Token Index: 265

Text #17

Max Range: 8.7104. Min Range: -8.7104

Max Act: 5.4782. Min Act: -0.0000

Data Index: 1962535 (The Pile)

Max Activating Token Index: 475

Text #18

Max Range: 8.7104. Min Range: -8.7104

Max Act: 5.3046. Min Act: -0.0000

Data Index: 1959110 (The Pile)

Max Activating Token Index: 311

Text #19

Max Range: 8.7104. Min Range: -8.7104

Max Act: 5.2548. Min Act: -0.0000

Data Index: 881872 (The Pile)

Max Activating Token Index: 840