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

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

Neuron 678 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.5412. Min Act: -0.0000

Data Index: 1187230 (C4 (Web Text))

Max Activating Token Index: 908

Text #1

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4670. Min Act: -0.0000

Data Index: 92948 (C4 (Web Text))

Max Activating Token Index: 766

Text #2

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4914. Min Act: -0.0000

Data Index: 961 (C4 (Web Text))

Max Activating Token Index: 334

Text #3

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4681. Min Act: -0.0001

Data Index: 1111539 (C4 (Web Text))

Max Activating Token Index: 376

Text #4

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4910. Min Act: -0.0001

Data Index: 405875 (C4 (Web Text))

Max Activating Token Index: 616

Text #5

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4320. Min Act: -0.0000

Data Index: 447817 (C4 (Web Text))

Max Activating Token Index: 298

Text #6

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4448. Min Act: -0.0001

Data Index: 753597 (C4 (Web Text))

Max Activating Token Index: 706

Text #7

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4578. Min Act: -0.0000

Data Index: 319387 (C4 (Web Text))

Max Activating Token Index: 653

Text #8

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4499. Min Act: -0.0001

Data Index: 290300 (C4 (Web Text))

Max Activating Token Index: 737

Text #9

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4482. Min Act: -0.0001

Data Index: 503108 (C4 (Web Text))

Max Activating Token Index: 52

Text #10

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4768. Min Act: -0.0000

Data Index: 772168 (C4 (Web Text))

Max Activating Token Index: 723

Text #11

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4361. Min Act: -0.0001

Data Index: 962887 (C4 (Web Text))

Max Activating Token Index: 34

Text #12

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.3763. Min Act: -0.0001

Data Index: 252323 (C4 (Web Text))

Max Activating Token Index: 965

Text #13

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4392. Min Act: -0.0001

Data Index: 857260 (C4 (Web Text))

Max Activating Token Index: 851

Text #14

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4010. Min Act: -0.0001

Data Index: 180513 (C4 (Web Text))

Max Activating Token Index: 401

Text #15

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4126. Min Act: -0.0001

Data Index: 1550140 (Python Code)

Max Activating Token Index: 834

Text #16

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.3921. Min Act: -0.0001

Data Index: 907703 (C4 (Web Text))

Max Activating Token Index: 638

Text #17

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.4149. Min Act: -0.0001

Data Index: 487337 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.3918. Min Act: -0.0000

Data Index: 689616 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 0.5412. Min Range: -0.5412

Max Act: 0.3785. Min Act: -0.0001

Data Index: 129070 (C4 (Web Text))

Max Activating Token Index: 231