Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 94 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0676. Min Act: -0.0003

Data Index: 950124 (C4 (Web Text))

Max Activating Token Index: 748

Text #1

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0717. Min Act: -0.0003

Data Index: 96868 (C4 (Web Text))

Max Activating Token Index: 639

Text #2

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0714. Min Act: -0.0003

Data Index: 1021974 (C4 (Web Text))

Max Activating Token Index: 861

Text #3

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0663. Min Act: -0.0003

Data Index: 912384 (C4 (Web Text))

Max Activating Token Index: 168

Text #4

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0689. Min Act: -0.0003

Data Index: 1318383 (C4 (Web Text))

Max Activating Token Index: 118

Text #5

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0623. Min Act: -0.0003

Data Index: 690732 (C4 (Web Text))

Max Activating Token Index: 993

Text #6

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0606. Min Act: -0.0003

Data Index: 1205959 (C4 (Web Text))

Max Activating Token Index: 95

Text #7

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0629. Min Act: -0.0003

Data Index: 1419564 (Python Code)

Max Activating Token Index: 79

Text #8

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0610. Min Act: -0.0003

Data Index: 110284 (C4 (Web Text))

Max Activating Token Index: 187

Text #9

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0598. Min Act: -0.0003

Data Index: 751697 (C4 (Web Text))

Max Activating Token Index: 263

Text #10

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0631. Min Act: -0.0003

Data Index: 350296 (C4 (Web Text))

Max Activating Token Index: 118

Text #11

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0632. Min Act: -0.0003

Data Index: 881814 (C4 (Web Text))

Max Activating Token Index: 404

Text #12

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0604. Min Act: -0.0003

Data Index: 936647 (C4 (Web Text))

Max Activating Token Index: 599

Text #13

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0577. Min Act: -0.0003

Data Index: 1462671 (Python Code)

Max Activating Token Index: 66

Text #14

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0592. Min Act: -0.0003

Data Index: 41668 (C4 (Web Text))

Max Activating Token Index: 322

Text #15

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0605. Min Act: -0.0003

Data Index: 1037774 (C4 (Web Text))

Max Activating Token Index: 220

Text #16

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0602. Min Act: -0.0003

Data Index: 79286 (C4 (Web Text))

Max Activating Token Index: 27

Text #17

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0584. Min Act: -0.0003

Data Index: 1478212 (Python Code)

Max Activating Token Index: 275

Text #18

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0589. Min Act: -0.0003

Data Index: 871854 (C4 (Web Text))

Max Activating Token Index: 389

Text #19

Max Range: 0.0717. Min Range: -0.0717

Max Act: 0.0571. Min Act: -0.0003

Data Index: 1207001 (C4 (Web Text))

Max Activating Token Index: 244