Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 24 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1811. Min Act: -0.0001

Data Index: 665542 (C4 (Web Text))

Max Activating Token Index: 895

Text #1

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1784. Min Act: -0.0001

Data Index: 789780 (C4 (Web Text))

Max Activating Token Index: 47

Text #2

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1676. Min Act: -0.0001

Data Index: 50451 (C4 (Web Text))

Max Activating Token Index: 928

Text #3

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1751. Min Act: -0.0001

Data Index: 233689 (C4 (Web Text))

Max Activating Token Index: 610

Text #4

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1711. Min Act: -0.0001

Data Index: 383491 (C4 (Web Text))

Max Activating Token Index: 874

Text #5

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1759. Min Act: -0.0001

Data Index: 1476949 (Python Code)

Max Activating Token Index: 439

Text #6

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1694. Min Act: -0.0001

Data Index: 1175157 (C4 (Web Text))

Max Activating Token Index: 187

Text #7

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1588. Min Act: -0.0001

Data Index: 682008 (C4 (Web Text))

Max Activating Token Index: 499

Text #8

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1738. Min Act: -0.0001

Data Index: 906587 (C4 (Web Text))

Max Activating Token Index: 621

Text #9

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1666. Min Act: -0.0001

Data Index: 889969 (C4 (Web Text))

Max Activating Token Index: 232

Text #10

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1625. Min Act: -0.0001

Data Index: 582134 (C4 (Web Text))

Max Activating Token Index: 568

Text #11

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1645. Min Act: -0.0001

Data Index: 1039381 (C4 (Web Text))

Max Activating Token Index: 690

Text #12

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1454. Min Act: -0.0001

Data Index: 72840 (C4 (Web Text))

Max Activating Token Index: 304

Text #13

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1604. Min Act: -0.0001

Data Index: 1181055 (C4 (Web Text))

Max Activating Token Index: 21

Text #14

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1652. Min Act: -0.0001

Data Index: 526302 (C4 (Web Text))

Max Activating Token Index: 489

Text #15

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1753. Min Act: -0.0001

Data Index: 770571 (C4 (Web Text))

Max Activating Token Index: 451

Text #16

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1465. Min Act: -0.0001

Data Index: 1185741 (C4 (Web Text))

Max Activating Token Index: 396

Text #17

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1490. Min Act: -0.0001

Data Index: 358259 (C4 (Web Text))

Max Activating Token Index: 477

Text #18

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1439. Min Act: -0.0001

Data Index: 517265 (C4 (Web Text))

Max Activating Token Index: 579

Text #19

Max Range: 0.1811. Min Range: -0.1811

Max Act: 0.1456. Min Act: -0.0001

Data Index: 197192 (C4 (Web Text))

Max Activating Token Index: 533