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

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

Neuron 1615 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0721. Min Act: -0.0001

Data Index: 1120335 (C4 (Web Text))

Max Activating Token Index: 898

Text #1

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0722. Min Act: -0.0001

Data Index: 1014940 (C4 (Web Text))

Max Activating Token Index: 389

Text #2

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0655. Min Act: -0.0001

Data Index: 345611 (C4 (Web Text))

Max Activating Token Index: 917

Text #3

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0722. Min Act: -0.0001

Data Index: 1244998 (C4 (Web Text))

Max Activating Token Index: 699

Text #4

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0683. Min Act: -0.0001

Data Index: 1263447 (C4 (Web Text))

Max Activating Token Index: 197

Text #5

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0647. Min Act: -0.0001

Data Index: 196463 (C4 (Web Text))

Max Activating Token Index: 729

Text #6

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0537. Min Act: -0.0001

Data Index: 53876 (C4 (Web Text))

Max Activating Token Index: 627

Text #7

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0641. Min Act: -0.0001

Data Index: 809866 (C4 (Web Text))

Max Activating Token Index: 738

Text #8

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0637. Min Act: -0.0000

Data Index: 1548116 (Python Code)

Max Activating Token Index: 705

Text #9

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0616. Min Act: -0.0001

Data Index: 1495735 (Python Code)

Max Activating Token Index: 119

Text #10

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0614. Min Act: -0.0000

Data Index: 1511679 (Python Code)

Max Activating Token Index: 381

Text #11

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0589. Min Act: -0.0000

Data Index: 946723 (C4 (Web Text))

Max Activating Token Index: 604

Text #12

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0626. Min Act: -0.0002

Data Index: 399326 (C4 (Web Text))

Max Activating Token Index: 525

Text #13

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0583. Min Act: -0.0000

Data Index: 141589 (C4 (Web Text))

Max Activating Token Index: 157

Text #14

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0660. Min Act: -0.0001

Data Index: 43997 (C4 (Web Text))

Max Activating Token Index: 1012

Text #15

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0602. Min Act: -0.0001

Data Index: 523113 (C4 (Web Text))

Max Activating Token Index: 457

Text #16

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0592. Min Act: -0.0001

Data Index: 1502606 (Python Code)

Max Activating Token Index: 34

Text #17

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0561. Min Act: -0.0001

Data Index: 852282 (C4 (Web Text))

Max Activating Token Index: 737

Text #18

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0580. Min Act: -0.0002

Data Index: 802211 (C4 (Web Text))

Max Activating Token Index: 976

Text #19

Max Range: 0.0722. Min Range: -0.0722

Max Act: 0.0576. Min Act: -0.0002

Data Index: 1540679 (Python Code)

Max Activating Token Index: 80