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

Dataset: The Pile

Neuron 999 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: 7.4061. Min Range: -7.4061

Max Act: 7.4061. Min Act: -0.0000

Data Index: 1428511 (The Pile)

Max Activating Token Index: 657

Text #1

Max Range: 7.4061. Min Range: -7.4061

Max Act: 7.0819. Min Act: -0.0000

Data Index: 768975 (The Pile)

Max Activating Token Index: 426

Text #2

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.9977. Min Act: -0.0000

Data Index: 1364917 (The Pile)

Max Activating Token Index: 65

Text #3

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.7307. Min Act: -0.0000

Data Index: 1669996 (The Pile)

Max Activating Token Index: 253

Text #4

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.4519. Min Act: -0.0000

Data Index: 2357 (The Pile)

Max Activating Token Index: 53

Text #5

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.1105. Min Act: -0.0000

Data Index: 1985853 (The Pile)

Max Activating Token Index: 461

Text #6

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.1063. Min Act: -0.0000

Data Index: 271643 (The Pile)

Max Activating Token Index: 300

Text #7

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.1492. Min Act: -0.0000

Data Index: 1727338 (The Pile)

Max Activating Token Index: 72

Text #8

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.2889. Min Act: -0.0000

Data Index: 241778 (The Pile)

Max Activating Token Index: 545

Text #9

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.0617. Min Act: -0.0000

Data Index: 1995480 (The Pile)

Max Activating Token Index: 782

Text #10

Max Range: 7.4061. Min Range: -7.4061

Max Act: 6.2295. Min Act: -0.0000

Data Index: 627547 (The Pile)

Max Activating Token Index: 566

Text #11

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.7546. Min Act: -0.0000

Data Index: 1682559 (The Pile)

Max Activating Token Index: 630

Text #12

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.9496. Min Act: -0.0000

Data Index: 821622 (The Pile)

Max Activating Token Index: 67

Text #13

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.7960. Min Act: -0.0000

Data Index: 586997 (The Pile)

Max Activating Token Index: 243

Text #14

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.5566. Min Act: -0.0000

Data Index: 95614 (The Pile)

Max Activating Token Index: 392

Text #15

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.6949. Min Act: -0.0000

Data Index: 1130707 (The Pile)

Max Activating Token Index: 360

Text #16

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.5386. Min Act: -0.0000

Data Index: 376584 (The Pile)

Max Activating Token Index: 82

Text #17

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.5256. Min Act: -0.0000

Data Index: 737811 (The Pile)

Max Activating Token Index: 100

Text #18

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.5611. Min Act: -0.0000

Data Index: 646043 (The Pile)

Max Activating Token Index: 62

Text #19

Max Range: 7.4061. Min Range: -7.4061

Max Act: 5.5459. Min Act: -0.0000

Data Index: 352937 (The Pile)

Max Activating Token Index: 82