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

Dataset: The Pile

Neuron 1749 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.1723. Min Range: -4.1723

Max Act: 4.1723. Min Act: -0.0000

Data Index: 1563219 (The Pile)

Max Activating Token Index: 26

Text #1

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.7277. Min Act: -0.0000

Data Index: 795960 (The Pile)

Max Activating Token Index: 16

Text #2

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.4232. Min Act: -0.0000

Data Index: 779355 (The Pile)

Max Activating Token Index: 62

Text #3

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.4392. Min Act: 0.0000

Data Index: 850440 (The Pile)

Max Activating Token Index: 59

Text #4

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.2104. Min Act: -0.0000

Data Index: 1024451 (The Pile)

Max Activating Token Index: 388

Text #5

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.0684. Min Act: 0.0000

Data Index: 1215419 (The Pile)

Max Activating Token Index: 49

Text #6

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.9091. Min Act: -0.0000

Data Index: 1826828 (The Pile)

Max Activating Token Index: 905

Text #7

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.1436. Min Act: -0.0000

Data Index: 447073 (The Pile)

Max Activating Token Index: 149

Text #8

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.9055. Min Act: -0.0000

Data Index: 1690360 (The Pile)

Max Activating Token Index: 174

Text #9

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.0166. Min Act: -0.0000

Data Index: 750638 (The Pile)

Max Activating Token Index: 995

Text #10

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.9730. Min Act: -0.0000

Data Index: 1293290 (The Pile)

Max Activating Token Index: 367

Text #11

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.0877. Min Act: 0.0000

Data Index: 1476679 (The Pile)

Max Activating Token Index: 47

Text #12

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.1007. Min Act: -0.0000

Data Index: 1498377 (The Pile)

Max Activating Token Index: 16

Text #13

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.0880. Min Act: -0.0000

Data Index: 580558 (The Pile)

Max Activating Token Index: 804

Text #14

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.9387. Min Act: -0.0000

Data Index: 395580 (The Pile)

Max Activating Token Index: 547

Text #15

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.9444. Min Act: 0.0000

Data Index: 372110 (The Pile)

Max Activating Token Index: 20

Text #16

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.0990. Min Act: -0.0000

Data Index: 880286 (The Pile)

Max Activating Token Index: 239

Text #17

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.7499. Min Act: -0.0000

Data Index: 1898467 (The Pile)

Max Activating Token Index: 598

Text #18

Max Range: 4.1723. Min Range: -4.1723

Max Act: 3.1587. Min Act: -0.0000

Data Index: 1728348 (The Pile)

Max Activating Token Index: 8

Text #19

Max Range: 4.1723. Min Range: -4.1723

Max Act: 2.8883. Min Act: -0.0000

Data Index: 761463 (The Pile)

Max Activating Token Index: 9