Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2669 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2654. Min Act: -0.0002

Data Index: 771757 (C4 (Web Text))

Max Activating Token Index: 185

Text #1

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2904. Min Act: -0.0002

Data Index: 295782 (C4 (Web Text))

Max Activating Token Index: 584

Text #2

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.3023. Min Act: -0.0002

Data Index: 415400 (C4 (Web Text))

Max Activating Token Index: 87

Text #3

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2939. Min Act: -0.0002

Data Index: 367804 (C4 (Web Text))

Max Activating Token Index: 20

Text #4

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2788. Min Act: -0.0002

Data Index: 135866 (C4 (Web Text))

Max Activating Token Index: 602

Text #5

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.3075. Min Act: -0.0002

Data Index: 79971 (C4 (Web Text))

Max Activating Token Index: 172

Text #6

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2857. Min Act: -0.0002

Data Index: 210014 (C4 (Web Text))

Max Activating Token Index: 257

Text #7

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2702. Min Act: -0.0002

Data Index: 184640 (C4 (Web Text))

Max Activating Token Index: 1004

Text #8

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2825. Min Act: -0.0002

Data Index: 314434 (C4 (Web Text))

Max Activating Token Index: 225

Text #9

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2592. Min Act: -0.0002

Data Index: 1061699 (C4 (Web Text))

Max Activating Token Index: 376

Text #10

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2390. Min Act: -0.0002

Data Index: 488521 (C4 (Web Text))

Max Activating Token Index: 278

Text #11

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2218. Min Act: -0.0002

Data Index: 421306 (C4 (Web Text))

Max Activating Token Index: 357

Text #12

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2602. Min Act: -0.0003

Data Index: 611999 (C4 (Web Text))

Max Activating Token Index: 959

Text #13

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2764. Min Act: -0.0002

Data Index: 861793 (C4 (Web Text))

Max Activating Token Index: 10

Text #14

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2410. Min Act: -0.0002

Data Index: 1399387 (Python Code)

Max Activating Token Index: 263

Text #15

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2481. Min Act: -0.0002

Data Index: 1294350 (C4 (Web Text))

Max Activating Token Index: 14

Text #16

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2386. Min Act: -0.0002

Data Index: 1324034 (C4 (Web Text))

Max Activating Token Index: 554

Text #17

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2265. Min Act: -0.0002

Data Index: 33723 (C4 (Web Text))

Max Activating Token Index: 435

Text #18

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2737. Min Act: -0.0002

Data Index: 374152 (C4 (Web Text))

Max Activating Token Index: 31

Text #19

Max Range: 0.3075. Min Range: -0.3075

Max Act: 0.2382. Min Act: -0.0002

Data Index: 1005651 (C4 (Web Text))

Max Activating Token Index: 744