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

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

Neuron 350 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: 0.0654. Min Range: -0.0654

Max Act: 0.0638. Min Act: -0.0002

Data Index: 993341 (C4 (Web Text))

Max Activating Token Index: 505

Text #1

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0646. Min Act: -0.0002

Data Index: 557701 (C4 (Web Text))

Max Activating Token Index: 693

Text #2

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0654. Min Act: -0.0002

Data Index: 263145 (C4 (Web Text))

Max Activating Token Index: 428

Text #3

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0631. Min Act: -0.0002

Data Index: 1208478 (C4 (Web Text))

Max Activating Token Index: 6

Text #4

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0617. Min Act: -0.0002

Data Index: 875035 (C4 (Web Text))

Max Activating Token Index: 133

Text #5

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0625. Min Act: -0.0002

Data Index: 160603 (C4 (Web Text))

Max Activating Token Index: 73

Text #6

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0607. Min Act: -0.0002

Data Index: 360891 (C4 (Web Text))

Max Activating Token Index: 174

Text #7

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0596. Min Act: -0.0002

Data Index: 1196580 (C4 (Web Text))

Max Activating Token Index: 64

Text #8

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0603. Min Act: -0.0002

Data Index: 526882 (C4 (Web Text))

Max Activating Token Index: 15

Text #9

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0602. Min Act: -0.0002

Data Index: 1328261 (C4 (Web Text))

Max Activating Token Index: 788

Text #10

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0588. Min Act: -0.0001

Data Index: 983281 (C4 (Web Text))

Max Activating Token Index: 196

Text #11

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0582. Min Act: -0.0002

Data Index: 608400 (C4 (Web Text))

Max Activating Token Index: 57

Text #12

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0556. Min Act: -0.0002

Data Index: 795961 (C4 (Web Text))

Max Activating Token Index: 588

Text #13

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0576. Min Act: -0.0002

Data Index: 893345 (C4 (Web Text))

Max Activating Token Index: 348

Text #14

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0583. Min Act: -0.0001

Data Index: 1135744 (C4 (Web Text))

Max Activating Token Index: 575

Text #15

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0562. Min Act: -0.0002

Data Index: 936468 (C4 (Web Text))

Max Activating Token Index: 26

Text #16

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0560. Min Act: -0.0001

Data Index: 554616 (C4 (Web Text))

Max Activating Token Index: 515

Text #17

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0537. Min Act: -0.0002

Data Index: 898571 (C4 (Web Text))

Max Activating Token Index: 911

Text #18

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0535. Min Act: -0.0002

Data Index: 1026016 (C4 (Web Text))

Max Activating Token Index: 503

Text #19

Max Range: 0.0654. Min Range: -0.0654

Max Act: 0.0576. Min Act: -0.0002

Data Index: 809911 (C4 (Web Text))

Max Activating Token Index: 151