Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: The Pile

Neuron 2946 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l-pile')

Text #0

Max Range: 10.0914. Min Range: -10.0914

Max Act: 10.0914. Min Act: -0.0000

Data Index: 472157 (The Pile)

Max Activating Token Index: 44

Text #1

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.8368. Min Act: -0.0000

Data Index: 1484462 (The Pile)

Max Activating Token Index: 38

Text #2

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.6961. Min Act: -0.0000

Data Index: 1192250 (The Pile)

Max Activating Token Index: 984

Text #3

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.3300. Min Act: -0.0000

Data Index: 993162 (The Pile)

Max Activating Token Index: 122

Text #4

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.4825. Min Act: -0.0000

Data Index: 359588 (The Pile)

Max Activating Token Index: 896

Text #5

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.4242. Min Act: -0.0000

Data Index: 992884 (The Pile)

Max Activating Token Index: 665

Text #6

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.4023. Min Act: -0.0000

Data Index: 691046 (The Pile)

Max Activating Token Index: 185

Text #7

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.0387. Min Act: -0.0000

Data Index: 143673 (The Pile)

Max Activating Token Index: 98

Text #8

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.1291. Min Act: -0.0000

Data Index: 1993254 (The Pile)

Max Activating Token Index: 223

Text #9

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.1546. Min Act: -0.0000

Data Index: 1353341 (The Pile)

Max Activating Token Index: 586

Text #10

Max Range: 10.0914. Min Range: -10.0914

Max Act: 8.6110. Min Act: -0.0000

Data Index: 1982483 (The Pile)

Max Activating Token Index: 819

Text #11

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.1079. Min Act: -0.0000

Data Index: 1817314 (The Pile)

Max Activating Token Index: 98

Text #12

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.0943. Min Act: -0.0000

Data Index: 731372 (The Pile)

Max Activating Token Index: 402

Text #13

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.2933. Min Act: -0.0000

Data Index: 1510067 (The Pile)

Max Activating Token Index: 443

Text #14

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.0822. Min Act: -0.0000

Data Index: 138886 (The Pile)

Max Activating Token Index: 607

Text #15

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.0920. Min Act: -0.0000

Data Index: 1146584 (The Pile)

Max Activating Token Index: 786

Text #16

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.0957. Min Act: -0.0000

Data Index: 1006364 (The Pile)

Max Activating Token Index: 407

Text #17

Max Range: 10.0914. Min Range: -10.0914

Max Act: 8.7462. Min Act: -0.0000

Data Index: 481414 (The Pile)

Max Activating Token Index: 616

Text #18

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.2001. Min Act: -0.0000

Data Index: 1500519 (The Pile)

Max Activating Token Index: 993

Text #19

Max Range: 10.0914. Min Range: -10.0914

Max Act: 9.0845. Min Act: -0.0000

Data Index: 773715 (The Pile)

Max Activating Token Index: 516