Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2730 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5399. Min Act: -0.0000

Data Index: 1598866 (Python Code)

Max Activating Token Index: 795

Text #1

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5688. Min Act: -0.0000

Data Index: 843677 (C4 (Web Text))

Max Activating Token Index: 520

Text #2

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5602. Min Act: -0.0000

Data Index: 5243 (C4 (Web Text))

Max Activating Token Index: 435

Text #3

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4916. Min Act: -0.0000

Data Index: 1401808 (Python Code)

Max Activating Token Index: 678

Text #4

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5077. Min Act: -0.0000

Data Index: 285087 (C4 (Web Text))

Max Activating Token Index: 356

Text #5

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5077. Min Act: -0.0000

Data Index: 914215 (C4 (Web Text))

Max Activating Token Index: 535

Text #6

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5136. Min Act: -0.0000

Data Index: 1424281 (Python Code)

Max Activating Token Index: 251

Text #7

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4786. Min Act: -0.0000

Data Index: 350574 (C4 (Web Text))

Max Activating Token Index: 672

Text #8

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4969. Min Act: -0.0000

Data Index: 1034133 (C4 (Web Text))

Max Activating Token Index: 576

Text #9

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4664. Min Act: -0.0000

Data Index: 1211610 (C4 (Web Text))

Max Activating Token Index: 423

Text #10

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4739. Min Act: -0.0000

Data Index: 1121642 (C4 (Web Text))

Max Activating Token Index: 88

Text #11

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5047. Min Act: -0.0000

Data Index: 423743 (C4 (Web Text))

Max Activating Token Index: 218

Text #12

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4708. Min Act: -0.0000

Data Index: 1132150 (C4 (Web Text))

Max Activating Token Index: 638

Text #13

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.5057. Min Act: -0.0000

Data Index: 525220 (C4 (Web Text))

Max Activating Token Index: 313

Text #14

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4605. Min Act: -0.0000

Data Index: 1566553 (Python Code)

Max Activating Token Index: 310

Text #15

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4694. Min Act: -0.0000

Data Index: 1074841 (C4 (Web Text))

Max Activating Token Index: 822

Text #16

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4575. Min Act: -0.0000

Data Index: 75980 (C4 (Web Text))

Max Activating Token Index: 123

Text #17

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4859. Min Act: -0.0000

Data Index: 850080 (C4 (Web Text))

Max Activating Token Index: 241

Text #18

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4554. Min Act: -0.0000

Data Index: 1625116 (Python Code)

Max Activating Token Index: 358

Text #19

Max Range: 0.5688. Min Range: -0.5688

Max Act: 0.4866. Min Act: -0.0000

Data Index: 230549 (C4 (Web Text))

Max Activating Token Index: 768