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

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3159 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1018. Min Range: -3.1018

Max Act: 3.0075. Min Act: -0.0000

Data Index: 788410 (C4 (Web Text))

Max Activating Token Index: 799

Text #1

Max Range: 3.1018. Min Range: -3.1018

Max Act: 3.0837. Min Act: -0.0000

Data Index: 899754 (C4 (Web Text))

Max Activating Token Index: 929

Text #2

Max Range: 3.1018. Min Range: -3.1018

Max Act: 3.1018. Min Act: -0.0000

Data Index: 51618 (C4 (Web Text))

Max Activating Token Index: 549

Text #3

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.8690. Min Act: -0.0000

Data Index: 616167 (C4 (Web Text))

Max Activating Token Index: 527

Text #4

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.5004. Min Act: -0.0000

Data Index: 1518330 (Python Code)

Max Activating Token Index: 131

Text #5

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.4608. Min Act: -0.0000

Data Index: 876037 (C4 (Web Text))

Max Activating Token Index: 1007

Text #6

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.5816. Min Act: -0.0000

Data Index: 740865 (C4 (Web Text))

Max Activating Token Index: 210

Text #7

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.1983. Min Act: -0.0000

Data Index: 49167 (C4 (Web Text))

Max Activating Token Index: 298

Text #8

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.8060. Min Act: -0.0000

Data Index: 606424 (C4 (Web Text))

Max Activating Token Index: 955

Text #9

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.3940. Min Act: -0.0000

Data Index: 1417085 (Python Code)

Max Activating Token Index: 375

Text #10

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.6567. Min Act: -0.0000

Data Index: 924902 (C4 (Web Text))

Max Activating Token Index: 467

Text #11

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.2327. Min Act: -0.0000

Data Index: 284909 (C4 (Web Text))

Max Activating Token Index: 728

Text #12

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.5848. Min Act: -0.0000

Data Index: 916711 (C4 (Web Text))

Max Activating Token Index: 635

Text #13

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.6703. Min Act: -0.0000

Data Index: 1621622 (Python Code)

Max Activating Token Index: 878

Text #14

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.2870. Min Act: -0.0000

Data Index: 388581 (C4 (Web Text))

Max Activating Token Index: 444

Text #15

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.4015. Min Act: -0.0000

Data Index: 110693 (C4 (Web Text))

Max Activating Token Index: 178

Text #16

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.5242. Min Act: -0.0000

Data Index: 749248 (C4 (Web Text))

Max Activating Token Index: 748

Text #17

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.3905. Min Act: -0.0000

Data Index: 973759 (C4 (Web Text))

Max Activating Token Index: 1012

Text #18

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.3632. Min Act: -0.0000

Data Index: 1210434 (C4 (Web Text))

Max Activating Token Index: 169

Text #19

Max Range: 3.1018. Min Range: -3.1018

Max Act: 2.4868. Min Act: -0.0000

Data Index: 891669 (C4 (Web Text))

Max Activating Token Index: 757