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

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

Neuron 3502 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0934. Min Act: -0.0001

Data Index: 82183 (C4 (Web Text))

Max Activating Token Index: 949

Text #1

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0922. Min Act: -0.0001

Data Index: 210593 (C4 (Web Text))

Max Activating Token Index: 998

Text #2

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0970. Min Act: -0.0002

Data Index: 770536 (C4 (Web Text))

Max Activating Token Index: 715

Text #3

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0917. Min Act: -0.0001

Data Index: 930253 (C4 (Web Text))

Max Activating Token Index: 334

Text #4

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0899. Min Act: -0.0001

Data Index: 117855 (C4 (Web Text))

Max Activating Token Index: 26

Text #5

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0834. Min Act: -0.0001

Data Index: 623508 (C4 (Web Text))

Max Activating Token Index: 815

Text #6

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0727. Min Act: -0.0002

Data Index: 837655 (C4 (Web Text))

Max Activating Token Index: 709

Text #7

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0852. Min Act: -0.0001

Data Index: 1304558 (C4 (Web Text))

Max Activating Token Index: 922

Text #8

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0878. Min Act: -0.0001

Data Index: 1224210 (C4 (Web Text))

Max Activating Token Index: 928

Text #9

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0813. Min Act: -0.0002

Data Index: 23853 (C4 (Web Text))

Max Activating Token Index: 12

Text #10

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0874. Min Act: -0.0001

Data Index: 333785 (C4 (Web Text))

Max Activating Token Index: 124

Text #11

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0790. Min Act: -0.0001

Data Index: 363954 (C4 (Web Text))

Max Activating Token Index: 142

Text #12

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0821. Min Act: -0.0001

Data Index: 590582 (C4 (Web Text))

Max Activating Token Index: 22

Text #13

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0791. Min Act: -0.0001

Data Index: 689361 (C4 (Web Text))

Max Activating Token Index: 1018

Text #14

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0877. Min Act: -0.0001

Data Index: 343369 (C4 (Web Text))

Max Activating Token Index: 604

Text #15

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0792. Min Act: -0.0001

Data Index: 992265 (C4 (Web Text))

Max Activating Token Index: 755

Text #16

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0814. Min Act: -0.0001

Data Index: 17986 (C4 (Web Text))

Max Activating Token Index: 625

Text #17

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0776. Min Act: -0.0001

Data Index: 1084207 (C4 (Web Text))

Max Activating Token Index: 507

Text #18

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0807. Min Act: -0.0001

Data Index: 694952 (C4 (Web Text))

Max Activating Token Index: 799

Text #19

Max Range: 0.0970. Min Range: -0.0970

Max Act: 0.0808. Min Act: -0.0002

Data Index: 728824 (C4 (Web Text))

Max Activating Token Index: 577