Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2887 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.8065. Min Act: -0.0001

Data Index: 812972 (C4 (Web Text))

Max Activating Token Index: 982

Text #1

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.3243. Min Act: -0.0001

Data Index: 927139 (C4 (Web Text))

Max Activating Token Index: 461

Text #2

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.1987. Min Act: -0.0001

Data Index: 809167 (C4 (Web Text))

Max Activating Token Index: 507

Text #3

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.2651. Min Act: -0.0000

Data Index: 356577 (C4 (Web Text))

Max Activating Token Index: 130

Text #4

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.4763. Min Act: -0.0000

Data Index: 577746 (C4 (Web Text))

Max Activating Token Index: 902

Text #5

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.9264. Min Act: -0.0000

Data Index: 44366 (C4 (Web Text))

Max Activating Token Index: 350

Text #6

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.0561. Min Act: -0.0001

Data Index: 828830 (C4 (Web Text))

Max Activating Token Index: 524

Text #7

Max Range: 3.8065. Min Range: -3.8065

Max Act: 3.0166. Min Act: -0.0001

Data Index: 1024363 (C4 (Web Text))

Max Activating Token Index: 105

Text #8

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.8862. Min Act: -0.0001

Data Index: 352961 (C4 (Web Text))

Max Activating Token Index: 105

Text #9

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.8482. Min Act: -0.0001

Data Index: 519519 (C4 (Web Text))

Max Activating Token Index: 577

Text #10

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.7322. Min Act: -0.0001

Data Index: 1188577 (C4 (Web Text))

Max Activating Token Index: 683

Text #11

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.8287. Min Act: -0.0001

Data Index: 313901 (C4 (Web Text))

Max Activating Token Index: 701

Text #12

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.8047. Min Act: -0.0001

Data Index: 710097 (C4 (Web Text))

Max Activating Token Index: 521

Text #13

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.9447. Min Act: -0.0000

Data Index: 904955 (C4 (Web Text))

Max Activating Token Index: 152

Text #14

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.7472. Min Act: -0.0001

Data Index: 47588 (C4 (Web Text))

Max Activating Token Index: 841

Text #15

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.7399. Min Act: -0.0001

Data Index: 439122 (C4 (Web Text))

Max Activating Token Index: 790

Text #16

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.7004. Min Act: -0.0001

Data Index: 1207551 (C4 (Web Text))

Max Activating Token Index: 580

Text #17

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.5681. Min Act: -0.0000

Data Index: 1127919 (C4 (Web Text))

Max Activating Token Index: 696

Text #18

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.6819. Min Act: -0.0000

Data Index: 748629 (C4 (Web Text))

Max Activating Token Index: 869

Text #19

Max Range: 3.8065. Min Range: -3.8065

Max Act: 2.6777. Min Act: -0.0000

Data Index: 446766 (C4 (Web Text))

Max Activating Token Index: 997