Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2006 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0999. Min Act: -0.0003

Data Index: 938191 (C4 (Web Text))

Max Activating Token Index: 550

Text #1

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0775. Min Act: -0.0003

Data Index: 1550215 (Python Code)

Max Activating Token Index: 327

Text #2

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0759. Min Act: -0.0003

Data Index: 280538 (C4 (Web Text))

Max Activating Token Index: 474

Text #3

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0787. Min Act: -0.0003

Data Index: 1049546 (C4 (Web Text))

Max Activating Token Index: 455

Text #4

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0723. Min Act: -0.0003

Data Index: 1377511 (Python Code)

Max Activating Token Index: 398

Text #5

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0749. Min Act: -0.0003

Data Index: 1125441 (C4 (Web Text))

Max Activating Token Index: 422

Text #6

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0759. Min Act: -0.0003

Data Index: 1355556 (C4 (Web Text))

Max Activating Token Index: 281

Text #7

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0780. Min Act: -0.0003

Data Index: 906163 (C4 (Web Text))

Max Activating Token Index: 304

Text #8

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0761. Min Act: -0.0003

Data Index: 811574 (C4 (Web Text))

Max Activating Token Index: 585

Text #9

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0728. Min Act: -0.0003

Data Index: 1544133 (Python Code)

Max Activating Token Index: 621

Text #10

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0782. Min Act: -0.0003

Data Index: 945544 (C4 (Web Text))

Max Activating Token Index: 504

Text #11

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0755. Min Act: -0.0003

Data Index: 187126 (C4 (Web Text))

Max Activating Token Index: 721

Text #12

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0749. Min Act: -0.0003

Data Index: 130253 (C4 (Web Text))

Max Activating Token Index: 818

Text #13

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0746. Min Act: -0.0003

Data Index: 455371 (C4 (Web Text))

Max Activating Token Index: 461

Text #14

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0738. Min Act: -0.0003

Data Index: 1280346 (C4 (Web Text))

Max Activating Token Index: 141

Text #15

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0714. Min Act: -0.0003

Data Index: 1538633 (Python Code)

Max Activating Token Index: 419

Text #16

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0638. Min Act: -0.0003

Data Index: 258508 (C4 (Web Text))

Max Activating Token Index: 843

Text #17

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0708. Min Act: -0.0003

Data Index: 181077 (C4 (Web Text))

Max Activating Token Index: 792

Text #18

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0660. Min Act: -0.0003

Data Index: 1593854 (Python Code)

Max Activating Token Index: 335

Text #19

Max Range: 0.0999. Min Range: -0.0999

Max Act: 0.0736. Min Act: -0.0003

Data Index: 369012 (C4 (Web Text))

Max Activating Token Index: 299