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

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

Neuron 4268 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.1585. Min Act: -0.0000

Data Index: 968503 (C4 (Web Text))

Max Activating Token Index: 970

Text #1

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.1244. Min Act: -0.0000

Data Index: 678543 (C4 (Web Text))

Max Activating Token Index: 618

Text #2

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.1219. Min Act: -0.0000

Data Index: 146787 (C4 (Web Text))

Max Activating Token Index: 939

Text #3

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.1135. Min Act: -0.0000

Data Index: 891262 (C4 (Web Text))

Max Activating Token Index: 104

Text #4

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.1012. Min Act: -0.0000

Data Index: 653772 (C4 (Web Text))

Max Activating Token Index: 785

Text #5

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.1101. Min Act: -0.0000

Data Index: 318756 (C4 (Web Text))

Max Activating Token Index: 963

Text #6

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0971. Min Act: -0.0000

Data Index: 954206 (C4 (Web Text))

Max Activating Token Index: 862

Text #7

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0997. Min Act: -0.0000

Data Index: 49358 (C4 (Web Text))

Max Activating Token Index: 213

Text #8

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0975. Min Act: -0.0000

Data Index: 688549 (C4 (Web Text))

Max Activating Token Index: 707

Text #9

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0964. Min Act: -0.0000

Data Index: 54759 (C4 (Web Text))

Max Activating Token Index: 630

Text #10

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0898. Min Act: -0.0000

Data Index: 741292 (C4 (Web Text))

Max Activating Token Index: 784

Text #11

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0848. Min Act: -0.0000

Data Index: 485922 (C4 (Web Text))

Max Activating Token Index: 880

Text #12

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0814. Min Act: -0.0000

Data Index: 255510 (C4 (Web Text))

Max Activating Token Index: 644

Text #13

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0793. Min Act: -0.0000

Data Index: 1438428 (Python Code)

Max Activating Token Index: 258

Text #14

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0830. Min Act: -0.0000

Data Index: 22570 (C4 (Web Text))

Max Activating Token Index: 910

Text #15

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0812. Min Act: -0.0000

Data Index: 490418 (C4 (Web Text))

Max Activating Token Index: 461

Text #16

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0865. Min Act: -0.0000

Data Index: 802086 (C4 (Web Text))

Max Activating Token Index: 892

Text #17

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0856. Min Act: -0.0000

Data Index: 254495 (C4 (Web Text))

Max Activating Token Index: 393

Text #18

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0765. Min Act: -0.0000

Data Index: 792541 (C4 (Web Text))

Max Activating Token Index: 792

Text #19

Max Range: 0.1585. Min Range: -0.1585

Max Act: 0.0772. Min Act: -0.0000

Data Index: 1340242 (C4 (Web Text))

Max Activating Token Index: 906