Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1297 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0481. Min Act: -0.0005

Data Index: 1588732 (Python Code)

Max Activating Token Index: 653

Text #1

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0450. Min Act: -0.0005

Data Index: 1076838 (C4 (Web Text))

Max Activating Token Index: 303

Text #2

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0424. Min Act: -0.0005

Data Index: 1056927 (C4 (Web Text))

Max Activating Token Index: 477

Text #3

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0409. Min Act: -0.0005

Data Index: 1587201 (Python Code)

Max Activating Token Index: 294

Text #4

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0380. Min Act: -0.0005

Data Index: 1358396 (C4 (Web Text))

Max Activating Token Index: 51

Text #5

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0387. Min Act: -0.0005

Data Index: 78794 (C4 (Web Text))

Max Activating Token Index: 308

Text #6

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0359. Min Act: -0.0005

Data Index: 932313 (C4 (Web Text))

Max Activating Token Index: 845

Text #7

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0350. Min Act: -0.0005

Data Index: 936486 (C4 (Web Text))

Max Activating Token Index: 640

Text #8

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0393. Min Act: -0.0005

Data Index: 1609380 (Python Code)

Max Activating Token Index: 288

Text #9

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0371. Min Act: -0.0005

Data Index: 142842 (C4 (Web Text))

Max Activating Token Index: 609

Text #10

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0368. Min Act: -0.0005

Data Index: 1628237 (Python Code)

Max Activating Token Index: 480

Text #11

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0355. Min Act: -0.0005

Data Index: 573805 (C4 (Web Text))

Max Activating Token Index: 434

Text #12

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0366. Min Act: -0.0005

Data Index: 1091576 (C4 (Web Text))

Max Activating Token Index: 234

Text #13

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0381. Min Act: -0.0005

Data Index: 737840 (C4 (Web Text))

Max Activating Token Index: 578

Text #14

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0371. Min Act: -0.0005

Data Index: 430504 (C4 (Web Text))

Max Activating Token Index: 489

Text #15

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0365. Min Act: -0.0005

Data Index: 881435 (C4 (Web Text))

Max Activating Token Index: 898

Text #16

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0354. Min Act: -0.0004

Data Index: 751840 (C4 (Web Text))

Max Activating Token Index: 978

Text #17

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0351. Min Act: -0.0005

Data Index: 268236 (C4 (Web Text))

Max Activating Token Index: 486

Text #18

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0346. Min Act: -0.0005

Data Index: 1553273 (Python Code)

Max Activating Token Index: 18

Text #19

Max Range: 0.0481. Min Range: -0.0481

Max Act: 0.0355. Min Act: -0.0005

Data Index: 1239421 (C4 (Web Text))

Max Activating Token Index: 858