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

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

Neuron 768 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: 0.0602. Min Range: -0.0602

Max Act: 0.0593. Min Act: -0.0005

Data Index: 1108173 (C4 (Web Text))

Max Activating Token Index: 1002

Text #1

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0602. Min Act: -0.0005

Data Index: 1225281 (C4 (Web Text))

Max Activating Token Index: 73

Text #2

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0595. Min Act: -0.0005

Data Index: 179801 (C4 (Web Text))

Max Activating Token Index: 686

Text #3

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0577. Min Act: -0.0005

Data Index: 401690 (C4 (Web Text))

Max Activating Token Index: 60

Text #4

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0568. Min Act: -0.0005

Data Index: 368039 (C4 (Web Text))

Max Activating Token Index: 551

Text #5

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0534. Min Act: -0.0005

Data Index: 192934 (C4 (Web Text))

Max Activating Token Index: 743

Text #6

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0533. Min Act: -0.0005

Data Index: 1605810 (Python Code)

Max Activating Token Index: 702

Text #7

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0544. Min Act: -0.0005

Data Index: 826775 (C4 (Web Text))

Max Activating Token Index: 570

Text #8

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0525. Min Act: -0.0005

Data Index: 1614196 (Python Code)

Max Activating Token Index: 501

Text #9

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0554. Min Act: -0.0005

Data Index: 215554 (C4 (Web Text))

Max Activating Token Index: 1005

Text #10

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0552. Min Act: -0.0005

Data Index: 746441 (C4 (Web Text))

Max Activating Token Index: 669

Text #11

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0501. Min Act: -0.0005

Data Index: 407754 (C4 (Web Text))

Max Activating Token Index: 630

Text #12

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0529. Min Act: -0.0005

Data Index: 1281233 (C4 (Web Text))

Max Activating Token Index: 666

Text #13

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0530. Min Act: -0.0005

Data Index: 759200 (C4 (Web Text))

Max Activating Token Index: 149

Text #14

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0528. Min Act: -0.0005

Data Index: 608408 (C4 (Web Text))

Max Activating Token Index: 820

Text #15

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0504. Min Act: -0.0005

Data Index: 1169456 (C4 (Web Text))

Max Activating Token Index: 159

Text #16

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0523. Min Act: -0.0005

Data Index: 928627 (C4 (Web Text))

Max Activating Token Index: 594

Text #17

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0501. Min Act: -0.0005

Data Index: 239680 (C4 (Web Text))

Max Activating Token Index: 201

Text #18

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0508. Min Act: -0.0005

Data Index: 209130 (C4 (Web Text))

Max Activating Token Index: 611

Text #19

Max Range: 0.0602. Min Range: -0.0602

Max Act: 0.0510. Min Act: -0.0005

Data Index: 530698 (C4 (Web Text))

Max Activating Token Index: 662