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

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

Neuron 994 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.4725. Min Act: -0.0001

Data Index: 1005243 (C4 (Web Text))

Max Activating Token Index: 471

Text #1

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.4414. Min Act: -0.0001

Data Index: 1106520 (C4 (Web Text))

Max Activating Token Index: 253

Text #2

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3856. Min Act: -0.0001

Data Index: 88589 (C4 (Web Text))

Max Activating Token Index: 1012

Text #3

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3602. Min Act: -0.0001

Data Index: 1413988 (Python Code)

Max Activating Token Index: 405

Text #4

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3632. Min Act: -0.0001

Data Index: 1195284 (C4 (Web Text))

Max Activating Token Index: 452

Text #5

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3560. Min Act: -0.0001

Data Index: 442025 (C4 (Web Text))

Max Activating Token Index: 304

Text #6

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3508. Min Act: -0.0001

Data Index: 482439 (C4 (Web Text))

Max Activating Token Index: 963

Text #7

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3589. Min Act: -0.0001

Data Index: 1516719 (Python Code)

Max Activating Token Index: 662

Text #8

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3711. Min Act: -0.0001

Data Index: 1405530 (Python Code)

Max Activating Token Index: 5

Text #9

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3702. Min Act: -0.0001

Data Index: 1084876 (C4 (Web Text))

Max Activating Token Index: 406

Text #10

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3522. Min Act: -0.0001

Data Index: 984178 (C4 (Web Text))

Max Activating Token Index: 294

Text #11

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3433. Min Act: -0.0001

Data Index: 1374122 (Python Code)

Max Activating Token Index: 634

Text #12

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3504. Min Act: -0.0001

Data Index: 1586037 (Python Code)

Max Activating Token Index: 709

Text #13

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3338. Min Act: -0.0001

Data Index: 742525 (C4 (Web Text))

Max Activating Token Index: 417

Text #14

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3346. Min Act: -0.0001

Data Index: 312375 (C4 (Web Text))

Max Activating Token Index: 265

Text #15

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3429. Min Act: -0.0001

Data Index: 846826 (C4 (Web Text))

Max Activating Token Index: 714

Text #16

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3282. Min Act: -0.0001

Data Index: 1443233 (Python Code)

Max Activating Token Index: 375

Text #17

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3522. Min Act: -0.0001

Data Index: 1474355 (Python Code)

Max Activating Token Index: 938

Text #18

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3400. Min Act: -0.0001

Data Index: 991667 (C4 (Web Text))

Max Activating Token Index: 397

Text #19

Max Range: 0.4725. Min Range: -0.4725

Max Act: 0.3251. Min Act: -0.0001

Data Index: 9828 (C4 (Web Text))

Max Activating Token Index: 740