Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 78 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2972. Min Act: -0.0001

Data Index: 500143 (C4 (Web Text))

Max Activating Token Index: 995

Text #1

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2722. Min Act: -0.0001

Data Index: 737823 (C4 (Web Text))

Max Activating Token Index: 773

Text #2

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2758. Min Act: -0.0001

Data Index: 14040 (C4 (Web Text))

Max Activating Token Index: 388

Text #3

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2166. Min Act: -0.0001

Data Index: 986441 (C4 (Web Text))

Max Activating Token Index: 739

Text #4

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2011. Min Act: -0.0001

Data Index: 495265 (C4 (Web Text))

Max Activating Token Index: 750

Text #5

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2155. Min Act: -0.0001

Data Index: 710745 (C4 (Web Text))

Max Activating Token Index: 816

Text #6

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2229. Min Act: -0.0001

Data Index: 940423 (C4 (Web Text))

Max Activating Token Index: 54

Text #7

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.2013. Min Act: -0.0001

Data Index: 706187 (C4 (Web Text))

Max Activating Token Index: 270

Text #8

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1758. Min Act: -0.0001

Data Index: 1155771 (C4 (Web Text))

Max Activating Token Index: 915

Text #9

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1981. Min Act: -0.0001

Data Index: 264595 (C4 (Web Text))

Max Activating Token Index: 649

Text #10

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1840. Min Act: -0.0001

Data Index: 145032 (C4 (Web Text))

Max Activating Token Index: 938

Text #11

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1712. Min Act: -0.0001

Data Index: 3374 (C4 (Web Text))

Max Activating Token Index: 713

Text #12

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1724. Min Act: -0.0001

Data Index: 176620 (C4 (Web Text))

Max Activating Token Index: 386

Text #13

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1648. Min Act: -0.0001

Data Index: 479305 (C4 (Web Text))

Max Activating Token Index: 188

Text #14

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1626. Min Act: -0.0001

Data Index: 822647 (C4 (Web Text))

Max Activating Token Index: 937

Text #15

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1570. Min Act: -0.0001

Data Index: 599350 (C4 (Web Text))

Max Activating Token Index: 63

Text #16

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1592. Min Act: -0.0001

Data Index: 51016 (C4 (Web Text))

Max Activating Token Index: 280

Text #17

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1594. Min Act: -0.0001

Data Index: 1301166 (C4 (Web Text))

Max Activating Token Index: 1008

Text #18

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1565. Min Act: -0.0001

Data Index: 1345775 (C4 (Web Text))

Max Activating Token Index: 54

Text #19

Max Range: 0.2972. Min Range: -0.2972

Max Act: 0.1407. Min Act: -0.0001

Data Index: 110640 (C4 (Web Text))

Max Activating Token Index: 259