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

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

Neuron 3385 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0590. Min Act: -0.0003

Data Index: 70444 (C4 (Web Text))

Max Activating Token Index: 454

Text #1

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0606. Min Act: -0.0003

Data Index: 1086555 (C4 (Web Text))

Max Activating Token Index: 584

Text #2

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0573. Min Act: -0.0003

Data Index: 249686 (C4 (Web Text))

Max Activating Token Index: 158

Text #3

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0569. Min Act: -0.0003

Data Index: 1222979 (C4 (Web Text))

Max Activating Token Index: 798

Text #4

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0545. Min Act: -0.0003

Data Index: 263842 (C4 (Web Text))

Max Activating Token Index: 868

Text #5

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0518. Min Act: -0.0003

Data Index: 624138 (C4 (Web Text))

Max Activating Token Index: 772

Text #6

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0512. Min Act: -0.0003

Data Index: 767454 (C4 (Web Text))

Max Activating Token Index: 341

Text #7

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0544. Min Act: -0.0003

Data Index: 508761 (C4 (Web Text))

Max Activating Token Index: 579

Text #8

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0518. Min Act: -0.0003

Data Index: 573488 (C4 (Web Text))

Max Activating Token Index: 297

Text #9

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0498. Min Act: -0.0003

Data Index: 1273923 (C4 (Web Text))

Max Activating Token Index: 363

Text #10

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0511. Min Act: -0.0003

Data Index: 443232 (C4 (Web Text))

Max Activating Token Index: 30

Text #11

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0501. Min Act: -0.0003

Data Index: 178741 (C4 (Web Text))

Max Activating Token Index: 516

Text #12

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0494. Min Act: -0.0003

Data Index: 1362914 (Python Code)

Max Activating Token Index: 431

Text #13

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0506. Min Act: -0.0003

Data Index: 1285973 (C4 (Web Text))

Max Activating Token Index: 407

Text #14

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0477. Min Act: -0.0003

Data Index: 76895 (C4 (Web Text))

Max Activating Token Index: 666

Text #15

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0486. Min Act: -0.0003

Data Index: 938036 (C4 (Web Text))

Max Activating Token Index: 513

Text #16

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0502. Min Act: -0.0003

Data Index: 336560 (C4 (Web Text))

Max Activating Token Index: 996

Text #17

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0441. Min Act: -0.0003

Data Index: 1587358 (Python Code)

Max Activating Token Index: 495

Text #18

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0450. Min Act: -0.0003

Data Index: 1198879 (C4 (Web Text))

Max Activating Token Index: 509

Text #19

Max Range: 0.0606. Min Range: -0.0606

Max Act: 0.0446. Min Act: -0.0003

Data Index: 346626 (C4 (Web Text))

Max Activating Token Index: 289