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

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

Neuron 3807 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3588. Min Act: -0.0000

Data Index: 343380 (C4 (Web Text))

Max Activating Token Index: 801

Text #1

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3666. Min Act: -0.0001

Data Index: 452787 (C4 (Web Text))

Max Activating Token Index: 793

Text #2

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3595. Min Act: -0.0001

Data Index: 1036706 (C4 (Web Text))

Max Activating Token Index: 697

Text #3

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3621. Min Act: -0.0001

Data Index: 969717 (C4 (Web Text))

Max Activating Token Index: 881

Text #4

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3223. Min Act: -0.0001

Data Index: 95855 (C4 (Web Text))

Max Activating Token Index: 864

Text #5

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3599. Min Act: -0.0000

Data Index: 478477 (C4 (Web Text))

Max Activating Token Index: 679

Text #6

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3358. Min Act: -0.0001

Data Index: 1067757 (C4 (Web Text))

Max Activating Token Index: 754

Text #7

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3419. Min Act: -0.0001

Data Index: 518099 (C4 (Web Text))

Max Activating Token Index: 805

Text #8

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3271. Min Act: -0.0000

Data Index: 983660 (C4 (Web Text))

Max Activating Token Index: 815

Text #9

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3385. Min Act: -0.0000

Data Index: 619758 (C4 (Web Text))

Max Activating Token Index: 852

Text #10

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3025. Min Act: -0.0000

Data Index: 633355 (C4 (Web Text))

Max Activating Token Index: 609

Text #11

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3370. Min Act: -0.0001

Data Index: 639014 (C4 (Web Text))

Max Activating Token Index: 850

Text #12

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3601. Min Act: -0.0001

Data Index: 415815 (C4 (Web Text))

Max Activating Token Index: 789

Text #13

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3130. Min Act: -0.0000

Data Index: 684911 (C4 (Web Text))

Max Activating Token Index: 664

Text #14

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3311. Min Act: -0.0001

Data Index: 457057 (C4 (Web Text))

Max Activating Token Index: 937

Text #15

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3391. Min Act: -0.0001

Data Index: 9343 (C4 (Web Text))

Max Activating Token Index: 660

Text #16

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3518. Min Act: -0.0001

Data Index: 970824 (C4 (Web Text))

Max Activating Token Index: 944

Text #17

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3484. Min Act: -0.0000

Data Index: 612422 (C4 (Web Text))

Max Activating Token Index: 803

Text #18

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3094. Min Act: -0.0000

Data Index: 754306 (C4 (Web Text))

Max Activating Token Index: 635

Text #19

Max Range: 0.3666. Min Range: -0.3666

Max Act: 0.3176. Min Act: -0.0001

Data Index: 485590 (C4 (Web Text))

Max Activating Token Index: 698