Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 777 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.1049. Min Act: -0.0002

Data Index: 867309 (C4 (Web Text))

Max Activating Token Index: 133

Text #1

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.1055. Min Act: -0.0002

Data Index: 953971 (C4 (Web Text))

Max Activating Token Index: 910

Text #2

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.1009. Min Act: -0.0002

Data Index: 278146 (C4 (Web Text))

Max Activating Token Index: 603

Text #3

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.1089. Min Act: -0.0002

Data Index: 624062 (C4 (Web Text))

Max Activating Token Index: 611

Text #4

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0952. Min Act: -0.0002

Data Index: 796619 (C4 (Web Text))

Max Activating Token Index: 241

Text #5

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.1013. Min Act: -0.0002

Data Index: 1320761 (C4 (Web Text))

Max Activating Token Index: 705

Text #6

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.1003. Min Act: -0.0002

Data Index: 1421 (C4 (Web Text))

Max Activating Token Index: 714

Text #7

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0976. Min Act: -0.0002

Data Index: 96761 (C4 (Web Text))

Max Activating Token Index: 73

Text #8

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0954. Min Act: -0.0002

Data Index: 206370 (C4 (Web Text))

Max Activating Token Index: 154

Text #9

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0951. Min Act: -0.0002

Data Index: 543967 (C4 (Web Text))

Max Activating Token Index: 280

Text #10

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0979. Min Act: -0.0002

Data Index: 24923 (C4 (Web Text))

Max Activating Token Index: 968

Text #11

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0949. Min Act: -0.0002

Data Index: 874318 (C4 (Web Text))

Max Activating Token Index: 623

Text #12

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0910. Min Act: -0.0002

Data Index: 776900 (C4 (Web Text))

Max Activating Token Index: 605

Text #13

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0941. Min Act: -0.0002

Data Index: 401112 (C4 (Web Text))

Max Activating Token Index: 433

Text #14

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0965. Min Act: -0.0002

Data Index: 1075087 (C4 (Web Text))

Max Activating Token Index: 944

Text #15

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0924. Min Act: -0.0002

Data Index: 581159 (C4 (Web Text))

Max Activating Token Index: 396

Text #16

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0940. Min Act: -0.0002

Data Index: 581218 (C4 (Web Text))

Max Activating Token Index: 824

Text #17

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0950. Min Act: -0.0002

Data Index: 630257 (C4 (Web Text))

Max Activating Token Index: 622

Text #18

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0905. Min Act: -0.0002

Data Index: 564422 (C4 (Web Text))

Max Activating Token Index: 763

Text #19

Max Range: 0.1089. Min Range: -0.1089

Max Act: 0.0930. Min Act: -0.0002

Data Index: 587848 (C4 (Web Text))

Max Activating Token Index: 821