Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 548 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1148. Min Act: -0.0002

Data Index: 836979 (C4 (Web Text))

Max Activating Token Index: 602

Text #1

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1142. Min Act: -0.0002

Data Index: 1040607 (C4 (Web Text))

Max Activating Token Index: 914

Text #2

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1048. Min Act: -0.0002

Data Index: 53434 (C4 (Web Text))

Max Activating Token Index: 895

Text #3

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1035. Min Act: -0.0002

Data Index: 1189689 (C4 (Web Text))

Max Activating Token Index: 130

Text #4

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1025. Min Act: -0.0002

Data Index: 391075 (C4 (Web Text))

Max Activating Token Index: 829

Text #5

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1010. Min Act: -0.0002

Data Index: 81277 (C4 (Web Text))

Max Activating Token Index: 314

Text #6

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0991. Min Act: -0.0002

Data Index: 1358433 (C4 (Web Text))

Max Activating Token Index: 561

Text #7

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.1024. Min Act: -0.0002

Data Index: 540031 (C4 (Web Text))

Max Activating Token Index: 739

Text #8

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0970. Min Act: -0.0002

Data Index: 239697 (C4 (Web Text))

Max Activating Token Index: 273

Text #9

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0967. Min Act: -0.0002

Data Index: 1623378 (Python Code)

Max Activating Token Index: 337

Text #10

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0977. Min Act: -0.0002

Data Index: 1202723 (C4 (Web Text))

Max Activating Token Index: 425

Text #11

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0924. Min Act: -0.0002

Data Index: 958751 (C4 (Web Text))

Max Activating Token Index: 678

Text #12

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0992. Min Act: -0.0002

Data Index: 1614653 (Python Code)

Max Activating Token Index: 151

Text #13

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0947. Min Act: -0.0002

Data Index: 1639073 (Python Code)

Max Activating Token Index: 620

Text #14

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0921. Min Act: -0.0001

Data Index: 1649980 (Python Code)

Max Activating Token Index: 663

Text #15

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0929. Min Act: -0.0002

Data Index: 1503426 (Python Code)

Max Activating Token Index: 653

Text #16

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0951. Min Act: -0.0002

Data Index: 1527926 (Python Code)

Max Activating Token Index: 850

Text #17

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0972. Min Act: -0.0002

Data Index: 1458184 (Python Code)

Max Activating Token Index: 844

Text #18

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0980. Min Act: -0.0002

Data Index: 220802 (C4 (Web Text))

Max Activating Token Index: 77

Text #19

Max Range: 0.1148. Min Range: -0.1148

Max Act: 0.0985. Min Act: -0.0002

Data Index: 1051934 (C4 (Web Text))

Max Activating Token Index: 520