Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 342 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0879. Min Act: -0.0006

Data Index: 280967 (C4 (Web Text))

Max Activating Token Index: 606

Text #1

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0864. Min Act: -0.0006

Data Index: 3452 (C4 (Web Text))

Max Activating Token Index: 38

Text #2

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0804. Min Act: -0.0006

Data Index: 974442 (C4 (Web Text))

Max Activating Token Index: 396

Text #3

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0796. Min Act: -0.0005

Data Index: 600413 (C4 (Web Text))

Max Activating Token Index: 957

Text #4

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0704. Min Act: -0.0006

Data Index: 1204443 (C4 (Web Text))

Max Activating Token Index: 689

Text #5

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0755. Min Act: -0.0006

Data Index: 114631 (C4 (Web Text))

Max Activating Token Index: 526

Text #6

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0731. Min Act: -0.0006

Data Index: 718025 (C4 (Web Text))

Max Activating Token Index: 278

Text #7

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0684. Min Act: -0.0006

Data Index: 297838 (C4 (Web Text))

Max Activating Token Index: 619

Text #8

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0717. Min Act: -0.0006

Data Index: 807463 (C4 (Web Text))

Max Activating Token Index: 966

Text #9

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0740. Min Act: -0.0006

Data Index: 506323 (C4 (Web Text))

Max Activating Token Index: 619

Text #10

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0692. Min Act: -0.0006

Data Index: 856842 (C4 (Web Text))

Max Activating Token Index: 697

Text #11

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0687. Min Act: -0.0006

Data Index: 671508 (C4 (Web Text))

Max Activating Token Index: 450

Text #12

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0648. Min Act: -0.0006

Data Index: 1250757 (C4 (Web Text))

Max Activating Token Index: 329

Text #13

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0644. Min Act: -0.0005

Data Index: 941704 (C4 (Web Text))

Max Activating Token Index: 474

Text #14

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0609. Min Act: -0.0006

Data Index: 844763 (C4 (Web Text))

Max Activating Token Index: 616

Text #15

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0602. Min Act: -0.0006

Data Index: 832078 (C4 (Web Text))

Max Activating Token Index: 397

Text #16

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0600. Min Act: -0.0006

Data Index: 251957 (C4 (Web Text))

Max Activating Token Index: 30

Text #17

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0630. Min Act: -0.0006

Data Index: 1145113 (C4 (Web Text))

Max Activating Token Index: 910

Text #18

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0623. Min Act: -0.0006

Data Index: 331658 (C4 (Web Text))

Max Activating Token Index: 317

Text #19

Max Range: 0.0879. Min Range: -0.0879

Max Act: 0.0596. Min Act: -0.0006

Data Index: 1192423 (C4 (Web Text))

Max Activating Token Index: 240