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

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

Neuron 3870 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0744. Min Act: -0.0001

Data Index: 1229899 (C4 (Web Text))

Max Activating Token Index: 341

Text #1

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0671. Min Act: -0.0002

Data Index: 955983 (C4 (Web Text))

Max Activating Token Index: 625

Text #2

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0682. Min Act: -0.0001

Data Index: 191483 (C4 (Web Text))

Max Activating Token Index: 894

Text #3

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0644. Min Act: -0.0001

Data Index: 1503450 (Python Code)

Max Activating Token Index: 536

Text #4

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0696. Min Act: -0.0001

Data Index: 282082 (C4 (Web Text))

Max Activating Token Index: 559

Text #5

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0632. Min Act: -0.0002

Data Index: 1113689 (C4 (Web Text))

Max Activating Token Index: 683

Text #6

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0668. Min Act: -0.0002

Data Index: 753732 (C4 (Web Text))

Max Activating Token Index: 624

Text #7

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0688. Min Act: -0.0001

Data Index: 496173 (C4 (Web Text))

Max Activating Token Index: 40

Text #8

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0671. Min Act: -0.0002

Data Index: 375604 (C4 (Web Text))

Max Activating Token Index: 799

Text #9

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0670. Min Act: -0.0001

Data Index: 971226 (C4 (Web Text))

Max Activating Token Index: 418

Text #10

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0648. Min Act: -0.0001

Data Index: 978474 (C4 (Web Text))

Max Activating Token Index: 4

Text #11

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0624. Min Act: -0.0001

Data Index: 885344 (C4 (Web Text))

Max Activating Token Index: 209

Text #12

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0647. Min Act: -0.0001

Data Index: 1656773 (Python Code)

Max Activating Token Index: 359

Text #13

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0635. Min Act: -0.0001

Data Index: 263159 (C4 (Web Text))

Max Activating Token Index: 168

Text #14

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0660. Min Act: -0.0001

Data Index: 1022161 (C4 (Web Text))

Max Activating Token Index: 293

Text #15

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0647. Min Act: -0.0002

Data Index: 356422 (C4 (Web Text))

Max Activating Token Index: 929

Text #16

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0629. Min Act: -0.0001

Data Index: 180881 (C4 (Web Text))

Max Activating Token Index: 645

Text #17

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0646. Min Act: -0.0001

Data Index: 508495 (C4 (Web Text))

Max Activating Token Index: 747

Text #18

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0603. Min Act: -0.0002

Data Index: 722018 (C4 (Web Text))

Max Activating Token Index: 292

Text #19

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0619. Min Act: -0.0001

Data Index: 743098 (C4 (Web Text))

Max Activating Token Index: 8