Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 641 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.8929. Min Act: -0.0002

Data Index: 1569812 (Python Code)

Max Activating Token Index: 971

Text #1

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.6985. Min Act: -0.0003

Data Index: 1177056 (C4 (Web Text))

Max Activating Token Index: 609

Text #2

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.6262. Min Act: -0.0003

Data Index: 1216604 (C4 (Web Text))

Max Activating Token Index: 479

Text #3

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.6237. Min Act: -0.0002

Data Index: 1196637 (C4 (Web Text))

Max Activating Token Index: 746

Text #4

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5775. Min Act: -0.0002

Data Index: 795378 (C4 (Web Text))

Max Activating Token Index: 982

Text #5

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5937. Min Act: -0.0003

Data Index: 632269 (C4 (Web Text))

Max Activating Token Index: 683

Text #6

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5555. Min Act: -0.0003

Data Index: 1357879 (C4 (Web Text))

Max Activating Token Index: 910

Text #7

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5690. Min Act: -0.0003

Data Index: 800104 (C4 (Web Text))

Max Activating Token Index: 714

Text #8

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5552. Min Act: -0.0003

Data Index: 441643 (C4 (Web Text))

Max Activating Token Index: 238

Text #9

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5551. Min Act: -0.0003

Data Index: 138917 (C4 (Web Text))

Max Activating Token Index: 782

Text #10

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5542. Min Act: -0.0003

Data Index: 1229676 (C4 (Web Text))

Max Activating Token Index: 123

Text #11

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5528. Min Act: -0.0003

Data Index: 913236 (C4 (Web Text))

Max Activating Token Index: 1005

Text #12

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5507. Min Act: -0.0002

Data Index: 1592718 (Python Code)

Max Activating Token Index: 4

Text #13

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5680. Min Act: -0.0003

Data Index: 769107 (C4 (Web Text))

Max Activating Token Index: 616

Text #14

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5664. Min Act: -0.0003

Data Index: 1000560 (C4 (Web Text))

Max Activating Token Index: 928

Text #15

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5472. Min Act: -0.0003

Data Index: 375230 (C4 (Web Text))

Max Activating Token Index: 592

Text #16

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5407. Min Act: -0.0003

Data Index: 459046 (C4 (Web Text))

Max Activating Token Index: 687

Text #17

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5370. Min Act: -0.0003

Data Index: 311552 (C4 (Web Text))

Max Activating Token Index: 778

Text #18

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5484. Min Act: -0.0002

Data Index: 394301 (C4 (Web Text))

Max Activating Token Index: 456

Text #19

Max Range: 0.8929. Min Range: -0.8929

Max Act: 0.5307. Min Act: -0.0003

Data Index: 130566 (C4 (Web Text))

Max Activating Token Index: 201