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

Dataset: The Pile

Neuron 620 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: 8.3909. Min Range: -8.3909

Max Act: 8.2882. Min Act: -0.0000

Data Index: 103697 (The Pile)

Max Activating Token Index: 265

Text #1

Max Range: 8.3909. Min Range: -8.3909

Max Act: 8.3909. Min Act: -0.0000

Data Index: 482219 (The Pile)

Max Activating Token Index: 630

Text #2

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.8269. Min Act: -0.0000

Data Index: 1495160 (The Pile)

Max Activating Token Index: 875

Text #3

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.9364. Min Act: -0.0000

Data Index: 319204 (The Pile)

Max Activating Token Index: 134

Text #4

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.3371. Min Act: -0.0000

Data Index: 1716959 (The Pile)

Max Activating Token Index: 98

Text #5

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.1214. Min Act: -0.0000

Data Index: 363144 (The Pile)

Max Activating Token Index: 80

Text #6

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.2982. Min Act: -0.0000

Data Index: 1016763 (The Pile)

Max Activating Token Index: 162

Text #7

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.2858. Min Act: -0.0000

Data Index: 1710798 (The Pile)

Max Activating Token Index: 262

Text #8

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.1117. Min Act: -0.0000

Data Index: 1106785 (The Pile)

Max Activating Token Index: 364

Text #9

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.0525. Min Act: -0.0000

Data Index: 73701 (The Pile)

Max Activating Token Index: 11

Text #10

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.2167. Min Act: -0.0000

Data Index: 1511930 (The Pile)

Max Activating Token Index: 78

Text #11

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.1777. Min Act: -0.0000

Data Index: 927996 (The Pile)

Max Activating Token Index: 149

Text #12

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.5162. Min Act: -0.0000

Data Index: 1741517 (The Pile)

Max Activating Token Index: 376

Text #13

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.7056. Min Act: -0.0000

Data Index: 459351 (The Pile)

Max Activating Token Index: 79

Text #14

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.6096. Min Act: -0.0000

Data Index: 1438617 (The Pile)

Max Activating Token Index: 11

Text #15

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.7631. Min Act: -0.0000

Data Index: 16764 (The Pile)

Max Activating Token Index: 513

Text #16

Max Range: 8.3909. Min Range: -8.3909

Max Act: 7.0619. Min Act: -0.0000

Data Index: 457672 (The Pile)

Max Activating Token Index: 188

Text #17

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.9074. Min Act: -0.0000

Data Index: 1594845 (The Pile)

Max Activating Token Index: 377

Text #18

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.5391. Min Act: -0.0000

Data Index: 1240441 (The Pile)

Max Activating Token Index: 53

Text #19

Max Range: 8.3909. Min Range: -8.3909

Max Act: 6.5259. Min Act: -0.0000

Data Index: 1601201 (The Pile)

Max Activating Token Index: 424