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

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

Neuron 3252 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.3072. Min Act: -0.0000

Data Index: 1590262 (Python Code)

Max Activating Token Index: 994

Text #1

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2807. Min Act: -0.0000

Data Index: 1470641 (Python Code)

Max Activating Token Index: 431

Text #2

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2342. Min Act: -0.0000

Data Index: 418547 (C4 (Web Text))

Max Activating Token Index: 799

Text #3

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1950. Min Act: -0.0000

Data Index: 209804 (C4 (Web Text))

Max Activating Token Index: 866

Text #4

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2116. Min Act: -0.0000

Data Index: 1438455 (Python Code)

Max Activating Token Index: 827

Text #5

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2051. Min Act: -0.0000

Data Index: 1069614 (C4 (Web Text))

Max Activating Token Index: 44

Text #6

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1988. Min Act: -0.0000

Data Index: 694567 (C4 (Web Text))

Max Activating Token Index: 587

Text #7

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2004. Min Act: -0.0000

Data Index: 1160281 (C4 (Web Text))

Max Activating Token Index: 775

Text #8

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1929. Min Act: -0.0000

Data Index: 828975 (C4 (Web Text))

Max Activating Token Index: 231

Text #9

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2094. Min Act: -0.0000

Data Index: 761573 (C4 (Web Text))

Max Activating Token Index: 657

Text #10

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.2017. Min Act: -0.0000

Data Index: 1563768 (Python Code)

Max Activating Token Index: 222

Text #11

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1997. Min Act: -0.0000

Data Index: 86814 (C4 (Web Text))

Max Activating Token Index: 38

Text #12

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1916. Min Act: -0.0000

Data Index: 1608771 (Python Code)

Max Activating Token Index: 870

Text #13

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1858. Min Act: -0.0000

Data Index: 1395106 (Python Code)

Max Activating Token Index: 26

Text #14

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1887. Min Act: -0.0000

Data Index: 1120930 (C4 (Web Text))

Max Activating Token Index: 147

Text #15

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1697. Min Act: -0.0001

Data Index: 767177 (C4 (Web Text))

Max Activating Token Index: 271

Text #16

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1700. Min Act: -0.0000

Data Index: 1468566 (Python Code)

Max Activating Token Index: 448

Text #17

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1872. Min Act: -0.0000

Data Index: 1284168 (C4 (Web Text))

Max Activating Token Index: 286

Text #18

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1752. Min Act: -0.0000

Data Index: 1029667 (C4 (Web Text))

Max Activating Token Index: 400

Text #19

Max Range: 0.3072. Min Range: -0.3072

Max Act: 0.1771. Min Act: -0.0000

Data Index: 412519 (C4 (Web Text))

Max Activating Token Index: 68