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

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

Neuron 1627 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.1069. Min Act: -0.0001

Data Index: 790696 (C4 (Web Text))

Max Activating Token Index: 891

Text #1

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.1052. Min Act: -0.0001

Data Index: 1229178 (C4 (Web Text))

Max Activating Token Index: 514

Text #2

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0883. Min Act: -0.0001

Data Index: 81959 (C4 (Web Text))

Max Activating Token Index: 213

Text #3

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0810. Min Act: -0.0001

Data Index: 1626217 (Python Code)

Max Activating Token Index: 82

Text #4

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0748. Min Act: -0.0001

Data Index: 132902 (C4 (Web Text))

Max Activating Token Index: 317

Text #5

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0718. Min Act: -0.0001

Data Index: 577580 (C4 (Web Text))

Max Activating Token Index: 510

Text #6

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0763. Min Act: -0.0001

Data Index: 710081 (C4 (Web Text))

Max Activating Token Index: 919

Text #7

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0754. Min Act: -0.0001

Data Index: 1370614 (Python Code)

Max Activating Token Index: 602

Text #8

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0741. Min Act: -0.0001

Data Index: 977063 (C4 (Web Text))

Max Activating Token Index: 510

Text #9

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0674. Min Act: -0.0001

Data Index: 1080200 (C4 (Web Text))

Max Activating Token Index: 987

Text #10

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0695. Min Act: -0.0001

Data Index: 346626 (C4 (Web Text))

Max Activating Token Index: 557

Text #11

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0650. Min Act: -0.0001

Data Index: 657888 (C4 (Web Text))

Max Activating Token Index: 49

Text #12

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0671. Min Act: -0.0001

Data Index: 40610 (C4 (Web Text))

Max Activating Token Index: 465

Text #13

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0694. Min Act: -0.0001

Data Index: 8239 (C4 (Web Text))

Max Activating Token Index: 702

Text #14

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0650. Min Act: -0.0001

Data Index: 433587 (C4 (Web Text))

Max Activating Token Index: 834

Text #15

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0673. Min Act: -0.0001

Data Index: 1401320 (Python Code)

Max Activating Token Index: 493

Text #16

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0668. Min Act: -0.0001

Data Index: 1298623 (C4 (Web Text))

Max Activating Token Index: 770

Text #17

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0634. Min Act: -0.0001

Data Index: 279237 (C4 (Web Text))

Max Activating Token Index: 921

Text #18

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0625. Min Act: -0.0001

Data Index: 1166262 (C4 (Web Text))

Max Activating Token Index: 649

Text #19

Max Range: 0.1069. Min Range: -0.1069

Max Act: 0.0634. Min Act: -0.0001

Data Index: 535750 (C4 (Web Text))

Max Activating Token Index: 467