Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 32 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.1286. Min Act: -0.0001

Data Index: 509522 (C4 (Web Text))

Max Activating Token Index: 145

Text #1

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0958. Min Act: -0.0001

Data Index: 1254243 (C4 (Web Text))

Max Activating Token Index: 213

Text #2

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0899. Min Act: -0.0001

Data Index: 747863 (C4 (Web Text))

Max Activating Token Index: 753

Text #3

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0927. Min Act: -0.0001

Data Index: 482435 (C4 (Web Text))

Max Activating Token Index: 605

Text #4

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0833. Min Act: -0.0001

Data Index: 386223 (C4 (Web Text))

Max Activating Token Index: 954

Text #5

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0816. Min Act: -0.0001

Data Index: 531127 (C4 (Web Text))

Max Activating Token Index: 780

Text #6

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0788. Min Act: -0.0001

Data Index: 1429550 (Python Code)

Max Activating Token Index: 219

Text #7

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0822. Min Act: -0.0001

Data Index: 875265 (C4 (Web Text))

Max Activating Token Index: 646

Text #8

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0802. Min Act: -0.0001

Data Index: 1535535 (Python Code)

Max Activating Token Index: 343

Text #9

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0753. Min Act: -0.0001

Data Index: 514675 (C4 (Web Text))

Max Activating Token Index: 117

Text #10

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0807. Min Act: -0.0001

Data Index: 1561773 (Python Code)

Max Activating Token Index: 136

Text #11

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0766. Min Act: -0.0001

Data Index: 440820 (C4 (Web Text))

Max Activating Token Index: 785

Text #12

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0771. Min Act: -0.0001

Data Index: 1504888 (Python Code)

Max Activating Token Index: 624

Text #13

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0799. Min Act: -0.0001

Data Index: 1110934 (C4 (Web Text))

Max Activating Token Index: 121

Text #14

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0655. Min Act: -0.0001

Data Index: 507446 (C4 (Web Text))

Max Activating Token Index: 732

Text #15

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0775. Min Act: -0.0001

Data Index: 963760 (C4 (Web Text))

Max Activating Token Index: 410

Text #16

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0714. Min Act: -0.0001

Data Index: 1012617 (C4 (Web Text))

Max Activating Token Index: 731

Text #17

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0773. Min Act: -0.0001

Data Index: 357887 (C4 (Web Text))

Max Activating Token Index: 655

Text #18

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0768. Min Act: -0.0001

Data Index: 979105 (C4 (Web Text))

Max Activating Token Index: 626

Text #19

Max Range: 0.1286. Min Range: -0.1286

Max Act: 0.0744. Min Act: -0.0001

Data Index: 356487 (C4 (Web Text))

Max Activating Token Index: 18