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

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

Neuron 615 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: 0.1031. Min Range: -0.1031

Max Act: 0.1031. Min Act: -0.0001

Data Index: 519809 (C4 (Web Text))

Max Activating Token Index: 778

Text #1

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0925. Min Act: -0.0002

Data Index: 335410 (C4 (Web Text))

Max Activating Token Index: 403

Text #2

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0961. Min Act: -0.0001

Data Index: 1110530 (C4 (Web Text))

Max Activating Token Index: 216

Text #3

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0940. Min Act: -0.0001

Data Index: 915003 (C4 (Web Text))

Max Activating Token Index: 174

Text #4

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.1002. Min Act: -0.0001

Data Index: 410874 (C4 (Web Text))

Max Activating Token Index: 98

Text #5

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0754. Min Act: -0.0001

Data Index: 752956 (C4 (Web Text))

Max Activating Token Index: 80

Text #6

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0825. Min Act: -0.0001

Data Index: 853939 (C4 (Web Text))

Max Activating Token Index: 428

Text #7

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0778. Min Act: -0.0001

Data Index: 915304 (C4 (Web Text))

Max Activating Token Index: 311

Text #8

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0730. Min Act: -0.0001

Data Index: 1287074 (C4 (Web Text))

Max Activating Token Index: 235

Text #9

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0811. Min Act: -0.0001

Data Index: 979929 (C4 (Web Text))

Max Activating Token Index: 52

Text #10

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0808. Min Act: -0.0001

Data Index: 247639 (C4 (Web Text))

Max Activating Token Index: 223

Text #11

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0786. Min Act: -0.0001

Data Index: 627396 (C4 (Web Text))

Max Activating Token Index: 306

Text #12

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0792. Min Act: -0.0001

Data Index: 889017 (C4 (Web Text))

Max Activating Token Index: 144

Text #13

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0739. Min Act: -0.0001

Data Index: 1038733 (C4 (Web Text))

Max Activating Token Index: 458

Text #14

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0750. Min Act: -0.0002

Data Index: 728990 (C4 (Web Text))

Max Activating Token Index: 102

Text #15

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0690. Min Act: -0.0001

Data Index: 104557 (C4 (Web Text))

Max Activating Token Index: 47

Text #16

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0708. Min Act: -0.0001

Data Index: 461273 (C4 (Web Text))

Max Activating Token Index: 623

Text #17

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0704. Min Act: -0.0001

Data Index: 1107155 (C4 (Web Text))

Max Activating Token Index: 978

Text #18

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0755. Min Act: -0.0001

Data Index: 190088 (C4 (Web Text))

Max Activating Token Index: 49

Text #19

Max Range: 0.1031. Min Range: -0.1031

Max Act: 0.0765. Min Act: -0.0001

Data Index: 765698 (C4 (Web Text))

Max Activating Token Index: 108