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

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

Neuron 615 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.2788. Min Act: -0.0000

Data Index: 1110825 (C4 (Web Text))

Max Activating Token Index: 54

Text #1

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.2210. Min Act: -0.0000

Data Index: 1598695 (Python Code)

Max Activating Token Index: 981

Text #2

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.2163. Min Act: -0.0001

Data Index: 88499 (C4 (Web Text))

Max Activating Token Index: 447

Text #3

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.2051. Min Act: -0.0000

Data Index: 198826 (C4 (Web Text))

Max Activating Token Index: 782

Text #4

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1885. Min Act: -0.0000

Data Index: 319269 (C4 (Web Text))

Max Activating Token Index: 555

Text #5

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1841. Min Act: -0.0000

Data Index: 502031 (C4 (Web Text))

Max Activating Token Index: 615

Text #6

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1766. Min Act: -0.0001

Data Index: 351471 (C4 (Web Text))

Max Activating Token Index: 931

Text #7

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1749. Min Act: -0.0001

Data Index: 1498237 (Python Code)

Max Activating Token Index: 460

Text #8

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1854. Min Act: -0.0000

Data Index: 388426 (C4 (Web Text))

Max Activating Token Index: 975

Text #9

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1814. Min Act: -0.0001

Data Index: 1151986 (C4 (Web Text))

Max Activating Token Index: 913

Text #10

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1789. Min Act: -0.0000

Data Index: 713448 (C4 (Web Text))

Max Activating Token Index: 623

Text #11

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1727. Min Act: -0.0000

Data Index: 413911 (C4 (Web Text))

Max Activating Token Index: 512

Text #12

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1789. Min Act: -0.0001

Data Index: 478859 (C4 (Web Text))

Max Activating Token Index: 419

Text #13

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1773. Min Act: -0.0000

Data Index: 77792 (C4 (Web Text))

Max Activating Token Index: 40

Text #14

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1819. Min Act: -0.0001

Data Index: 215802 (C4 (Web Text))

Max Activating Token Index: 596

Text #15

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1740. Min Act: -0.0000

Data Index: 803451 (C4 (Web Text))

Max Activating Token Index: 987

Text #16

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1741. Min Act: -0.0001

Data Index: 1364728 (Python Code)

Max Activating Token Index: 278

Text #17

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1742. Min Act: -0.0001

Data Index: 1059473 (C4 (Web Text))

Max Activating Token Index: 98

Text #18

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1735. Min Act: -0.0001

Data Index: 261932 (C4 (Web Text))

Max Activating Token Index: 645

Text #19

Max Range: 0.2788. Min Range: -0.2788

Max Act: 0.1727. Min Act: -0.0000

Data Index: 355967 (C4 (Web Text))

Max Activating Token Index: 230