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

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

Neuron 1868 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0772. Min Act: -0.0005

Data Index: 624106 (C4 (Web Text))

Max Activating Token Index: 169

Text #1

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0734. Min Act: -0.0005

Data Index: 815336 (C4 (Web Text))

Max Activating Token Index: 581

Text #2

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0670. Min Act: -0.0005

Data Index: 867248 (C4 (Web Text))

Max Activating Token Index: 580

Text #3

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0683. Min Act: -0.0006

Data Index: 888720 (C4 (Web Text))

Max Activating Token Index: 860

Text #4

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0668. Min Act: -0.0005

Data Index: 222185 (C4 (Web Text))

Max Activating Token Index: 322

Text #5

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0631. Min Act: -0.0005

Data Index: 427251 (C4 (Web Text))

Max Activating Token Index: 199

Text #6

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0655. Min Act: -0.0005

Data Index: 970898 (C4 (Web Text))

Max Activating Token Index: 94

Text #7

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0691. Min Act: -0.0005

Data Index: 1349785 (C4 (Web Text))

Max Activating Token Index: 479

Text #8

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0636. Min Act: -0.0005

Data Index: 527987 (C4 (Web Text))

Max Activating Token Index: 432

Text #9

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0660. Min Act: -0.0005

Data Index: 721035 (C4 (Web Text))

Max Activating Token Index: 457

Text #10

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0623. Min Act: -0.0005

Data Index: 641200 (C4 (Web Text))

Max Activating Token Index: 21

Text #11

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0646. Min Act: -0.0005

Data Index: 491182 (C4 (Web Text))

Max Activating Token Index: 447

Text #12

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0600. Min Act: -0.0005

Data Index: 1243296 (C4 (Web Text))

Max Activating Token Index: 147

Text #13

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0571. Min Act: -0.0005

Data Index: 698013 (C4 (Web Text))

Max Activating Token Index: 227

Text #14

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0584. Min Act: -0.0005

Data Index: 387760 (C4 (Web Text))

Max Activating Token Index: 943

Text #15

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0582. Min Act: -0.0005

Data Index: 45821 (C4 (Web Text))

Max Activating Token Index: 134

Text #16

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0616. Min Act: -0.0005

Data Index: 916282 (C4 (Web Text))

Max Activating Token Index: 265

Text #17

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0617. Min Act: -0.0005

Data Index: 460054 (C4 (Web Text))

Max Activating Token Index: 793

Text #18

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0546. Min Act: -0.0005

Data Index: 365952 (C4 (Web Text))

Max Activating Token Index: 716

Text #19

Max Range: 0.0772. Min Range: -0.0772

Max Act: 0.0583. Min Act: -0.0005

Data Index: 1600584 (Python Code)

Max Activating Token Index: 411