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

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

Neuron 1365 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0471. Min Act: -0.0001

Data Index: 1004752 (C4 (Web Text))

Max Activating Token Index: 568

Text #1

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0506. Min Act: -0.0002

Data Index: 294387 (C4 (Web Text))

Max Activating Token Index: 364

Text #2

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0432. Min Act: -0.0002

Data Index: 983464 (C4 (Web Text))

Max Activating Token Index: 589

Text #3

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0414. Min Act: -0.0002

Data Index: 1525285 (Python Code)

Max Activating Token Index: 911

Text #4

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0426. Min Act: -0.0001

Data Index: 84257 (C4 (Web Text))

Max Activating Token Index: 464

Text #5

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0425. Min Act: -0.0002

Data Index: 888039 (C4 (Web Text))

Max Activating Token Index: 342

Text #6

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0424. Min Act: -0.0001

Data Index: 726425 (C4 (Web Text))

Max Activating Token Index: 865

Text #7

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0406. Min Act: -0.0002

Data Index: 1385499 (Python Code)

Max Activating Token Index: 723

Text #8

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0389. Min Act: -0.0002

Data Index: 1565275 (Python Code)

Max Activating Token Index: 788

Text #9

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0375. Min Act: -0.0002

Data Index: 128090 (C4 (Web Text))

Max Activating Token Index: 842

Text #10

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0384. Min Act: -0.0001

Data Index: 360540 (C4 (Web Text))

Max Activating Token Index: 358

Text #11

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0383. Min Act: -0.0002

Data Index: 1551998 (Python Code)

Max Activating Token Index: 469

Text #12

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0373. Min Act: -0.0002

Data Index: 1612009 (Python Code)

Max Activating Token Index: 727

Text #13

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0392. Min Act: -0.0002

Data Index: 168706 (C4 (Web Text))

Max Activating Token Index: 882

Text #14

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0387. Min Act: -0.0002

Data Index: 1623586 (Python Code)

Max Activating Token Index: 70

Text #15

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0377. Min Act: -0.0002

Data Index: 1583089 (Python Code)

Max Activating Token Index: 263

Text #16

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0366. Min Act: -0.0002

Data Index: 1648897 (Python Code)

Max Activating Token Index: 637

Text #17

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0358. Min Act: -0.0002

Data Index: 612452 (C4 (Web Text))

Max Activating Token Index: 876

Text #18

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0358. Min Act: -0.0002

Data Index: 1596531 (Python Code)

Max Activating Token Index: 998

Text #19

Max Range: 0.0506. Min Range: -0.0506

Max Act: 0.0353. Min Act: -0.0002

Data Index: 1391351 (Python Code)

Max Activating Token Index: 304