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

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

Neuron 4551 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.9558. Min Act: -0.0000

Data Index: 493898 (C4 (Web Text))

Max Activating Token Index: 905

Text #1

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.9023. Min Act: -0.0000

Data Index: 1100915 (C4 (Web Text))

Max Activating Token Index: 44

Text #2

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.8585. Min Act: -0.0000

Data Index: 165294 (C4 (Web Text))

Max Activating Token Index: 530

Text #3

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.8406. Min Act: -0.0000

Data Index: 913317 (C4 (Web Text))

Max Activating Token Index: 940

Text #4

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.8007. Min Act: -0.0000

Data Index: 1254156 (C4 (Web Text))

Max Activating Token Index: 337

Text #5

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.7788. Min Act: 0.0000

Data Index: 400174 (C4 (Web Text))

Max Activating Token Index: 1016

Text #6

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.7289. Min Act: -0.0000

Data Index: 462854 (C4 (Web Text))

Max Activating Token Index: 512

Text #7

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.7480. Min Act: -0.0000

Data Index: 833028 (C4 (Web Text))

Max Activating Token Index: 252

Text #8

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6691. Min Act: -0.0000

Data Index: 1298990 (C4 (Web Text))

Max Activating Token Index: 494

Text #9

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.7416. Min Act: -0.0000

Data Index: 472611 (C4 (Web Text))

Max Activating Token Index: 773

Text #10

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6952. Min Act: -0.0000

Data Index: 240616 (C4 (Web Text))

Max Activating Token Index: 313

Text #11

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6810. Min Act: 0.0001

Data Index: 1027892 (C4 (Web Text))

Max Activating Token Index: 729

Text #12

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6794. Min Act: -0.0000

Data Index: 520836 (C4 (Web Text))

Max Activating Token Index: 842

Text #13

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6517. Min Act: -0.0000

Data Index: 690486 (C4 (Web Text))

Max Activating Token Index: 534

Text #14

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6400. Min Act: -0.0000

Data Index: 862661 (C4 (Web Text))

Max Activating Token Index: 411

Text #15

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6527. Min Act: -0.0000

Data Index: 1379539 (Python Code)

Max Activating Token Index: 46

Text #16

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6478. Min Act: -0.0000

Data Index: 1141033 (C4 (Web Text))

Max Activating Token Index: 96

Text #17

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6253. Min Act: -0.0000

Data Index: 178824 (C4 (Web Text))

Max Activating Token Index: 886

Text #18

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6113. Min Act: -0.0000

Data Index: 104194 (C4 (Web Text))

Max Activating Token Index: 614

Text #19

Max Range: 0.9558. Min Range: -0.9558

Max Act: 0.6817. Min Act: -0.0000

Data Index: 144220 (C4 (Web Text))

Max Activating Token Index: 590