Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4620 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.2578. Min Range: -1.2578

Max Act: 1.2578. Min Act: -0.0000

Data Index: 65674 (C4 (Web Text))

Max Activating Token Index: 940

Text #1

Max Range: 1.2578. Min Range: -1.2578

Max Act: 1.1607. Min Act: -0.0001

Data Index: 1192786 (C4 (Web Text))

Max Activating Token Index: 865

Text #2

Max Range: 1.2578. Min Range: -1.2578

Max Act: 1.0860. Min Act: -0.0001

Data Index: 1338163 (C4 (Web Text))

Max Activating Token Index: 400

Text #3

Max Range: 1.2578. Min Range: -1.2578

Max Act: 1.0355. Min Act: -0.0000

Data Index: 1410581 (Python Code)

Max Activating Token Index: 22

Text #4

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.9690. Min Act: -0.0001

Data Index: 1629162 (Python Code)

Max Activating Token Index: 48

Text #5

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.9615. Min Act: -0.0001

Data Index: 1534413 (Python Code)

Max Activating Token Index: 37

Text #6

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.9421. Min Act: -0.0001

Data Index: 558118 (C4 (Web Text))

Max Activating Token Index: 302

Text #7

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.9033. Min Act: -0.0001

Data Index: 1006303 (C4 (Web Text))

Max Activating Token Index: 302

Text #8

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8400. Min Act: -0.0001

Data Index: 548927 (C4 (Web Text))

Max Activating Token Index: 522

Text #9

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8264. Min Act: -0.0001

Data Index: 1552632 (Python Code)

Max Activating Token Index: 30

Text #10

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8662. Min Act: -0.0001

Data Index: 1469730 (Python Code)

Max Activating Token Index: 606

Text #11

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8734. Min Act: -0.0000

Data Index: 673009 (C4 (Web Text))

Max Activating Token Index: 467

Text #12

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.9619. Min Act: -0.0000

Data Index: 298523 (C4 (Web Text))

Max Activating Token Index: 821

Text #13

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8666. Min Act: -0.0000

Data Index: 1250273 (C4 (Web Text))

Max Activating Token Index: 130

Text #14

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8683. Min Act: -0.0000

Data Index: 518445 (C4 (Web Text))

Max Activating Token Index: 742

Text #15

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8901. Min Act: -0.0001

Data Index: 1615862 (Python Code)

Max Activating Token Index: 29

Text #16

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.9034. Min Act: -0.0001

Data Index: 693976 (C4 (Web Text))

Max Activating Token Index: 323

Text #17

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8385. Min Act: -0.0001

Data Index: 478063 (C4 (Web Text))

Max Activating Token Index: 121

Text #18

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8824. Min Act: -0.0001

Data Index: 11964 (C4 (Web Text))

Max Activating Token Index: 960

Text #19

Max Range: 1.2578. Min Range: -1.2578

Max Act: 0.8144. Min Act: -0.0001

Data Index: 1091732 (C4 (Web Text))

Max Activating Token Index: 909