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

Dataset: The Pile

Neuron 2811 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 10.1940. Min Range: -10.1940

Max Act: 10.1940. Min Act: 0.0002

Data Index: 1878634 (The Pile)

Max Activating Token Index: 4

Text #1

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.6538. Min Act: -0.0000

Data Index: 375479 (The Pile)

Max Activating Token Index: 412

Text #2

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.6170. Min Act: -0.0000

Data Index: 1367142 (The Pile)

Max Activating Token Index: 903

Text #3

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.4613. Min Act: -0.0000

Data Index: 1823745 (The Pile)

Max Activating Token Index: 421

Text #4

Max Range: 10.1940. Min Range: -10.1940

Max Act: 10.0790. Min Act: -0.0000

Data Index: 1780400 (The Pile)

Max Activating Token Index: 917

Text #5

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.4990. Min Act: 0.0001

Data Index: 1537621 (The Pile)

Max Activating Token Index: 66

Text #6

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.2656. Min Act: -0.0000

Data Index: 163403 (The Pile)

Max Activating Token Index: 818

Text #7

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.2722. Min Act: -0.0000

Data Index: 187910 (The Pile)

Max Activating Token Index: 153

Text #8

Max Range: 10.1940. Min Range: -10.1940

Max Act: 9.0047. Min Act: -0.0000

Data Index: 900537 (The Pile)

Max Activating Token Index: 478

Text #9

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.9758. Min Act: -0.0000

Data Index: 545342 (The Pile)

Max Activating Token Index: 136

Text #10

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.9008. Min Act: -0.0000

Data Index: 616939 (The Pile)

Max Activating Token Index: 908

Text #11

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.8955. Min Act: -0.0000

Data Index: 1057152 (The Pile)

Max Activating Token Index: 346

Text #12

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.8543. Min Act: 0.0002

Data Index: 245426 (The Pile)

Max Activating Token Index: 12

Text #13

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.8707. Min Act: -0.0000

Data Index: 1033867 (The Pile)

Max Activating Token Index: 24

Text #14

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.8550. Min Act: -0.0000

Data Index: 1335513 (The Pile)

Max Activating Token Index: 450

Text #15

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.6980. Min Act: -0.0000

Data Index: 1097431 (The Pile)

Max Activating Token Index: 878

Text #16

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.6905. Min Act: -0.0000

Data Index: 926209 (The Pile)

Max Activating Token Index: 743

Text #17

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.4839. Min Act: 0.0002

Data Index: 1185678 (The Pile)

Max Activating Token Index: 39

Text #18

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.8141. Min Act: 0.0002

Data Index: 1758127 (The Pile)

Max Activating Token Index: 24

Text #19

Max Range: 10.1940. Min Range: -10.1940

Max Act: 8.7623. Min Act: -0.0000

Data Index: 310167 (The Pile)

Max Activating Token Index: 971