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

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

Neuron 289 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0666. Min Act: -0.0003

Data Index: 238700 (C4 (Web Text))

Max Activating Token Index: 561

Text #1

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0666. Min Act: -0.0004

Data Index: 308034 (C4 (Web Text))

Max Activating Token Index: 189

Text #2

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0632. Min Act: -0.0003

Data Index: 50596 (C4 (Web Text))

Max Activating Token Index: 632

Text #3

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0647. Min Act: -0.0003

Data Index: 131688 (C4 (Web Text))

Max Activating Token Index: 312

Text #4

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0652. Min Act: -0.0003

Data Index: 1117141 (C4 (Web Text))

Max Activating Token Index: 249

Text #5

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0647. Min Act: -0.0003

Data Index: 118310 (C4 (Web Text))

Max Activating Token Index: 575

Text #6

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0634. Min Act: -0.0003

Data Index: 1029806 (C4 (Web Text))

Max Activating Token Index: 694

Text #7

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0643. Min Act: -0.0003

Data Index: 804656 (C4 (Web Text))

Max Activating Token Index: 950

Text #8

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0627. Min Act: -0.0003

Data Index: 303282 (C4 (Web Text))

Max Activating Token Index: 717

Text #9

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0645. Min Act: -0.0003

Data Index: 32393 (C4 (Web Text))

Max Activating Token Index: 269

Text #10

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0615. Min Act: -0.0003

Data Index: 227015 (C4 (Web Text))

Max Activating Token Index: 265

Text #11

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0614. Min Act: -0.0003

Data Index: 1051283 (C4 (Web Text))

Max Activating Token Index: 262

Text #12

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0630. Min Act: -0.0003

Data Index: 641357 (C4 (Web Text))

Max Activating Token Index: 234

Text #13

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0641. Min Act: -0.0004

Data Index: 983533 (C4 (Web Text))

Max Activating Token Index: 791

Text #14

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0604. Min Act: -0.0003

Data Index: 1392237 (Python Code)

Max Activating Token Index: 566

Text #15

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0617. Min Act: -0.0003

Data Index: 489779 (C4 (Web Text))

Max Activating Token Index: 369

Text #16

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0603. Min Act: -0.0003

Data Index: 1619457 (Python Code)

Max Activating Token Index: 574

Text #17

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0640. Min Act: -0.0003

Data Index: 1248961 (C4 (Web Text))

Max Activating Token Index: 461

Text #18

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0602. Min Act: -0.0003

Data Index: 565505 (C4 (Web Text))

Max Activating Token Index: 780

Text #19

Max Range: 0.0666. Min Range: -0.0666

Max Act: 0.0629. Min Act: -0.0003

Data Index: 1231325 (C4 (Web Text))

Max Activating Token Index: 583