Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 218 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 1.0867. Min Range: -1.0867

Max Act: 1.0867. Min Act: -0.0003

Data Index: 436564 (C4 (Web Text))

Max Activating Token Index: 592

Text #1

Max Range: 1.0867. Min Range: -1.0867

Max Act: 1.0364. Min Act: -0.0002

Data Index: 894921 (C4 (Web Text))

Max Activating Token Index: 147

Text #2

Max Range: 1.0867. Min Range: -1.0867

Max Act: 1.0006. Min Act: -0.0002

Data Index: 228279 (C4 (Web Text))

Max Activating Token Index: 734

Text #3

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9526. Min Act: -0.0002

Data Index: 489592 (C4 (Web Text))

Max Activating Token Index: 275

Text #4

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9733. Min Act: -0.0002

Data Index: 418498 (C4 (Web Text))

Max Activating Token Index: 479

Text #5

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9284. Min Act: -0.0002

Data Index: 832475 (C4 (Web Text))

Max Activating Token Index: 734

Text #6

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9421. Min Act: -0.0002

Data Index: 823309 (C4 (Web Text))

Max Activating Token Index: 14

Text #7

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9313. Min Act: -0.0002

Data Index: 815977 (C4 (Web Text))

Max Activating Token Index: 719

Text #8

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9210. Min Act: -0.0002

Data Index: 950878 (C4 (Web Text))

Max Activating Token Index: 811

Text #9

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9096. Min Act: -0.0002

Data Index: 1359424 (C4 (Web Text))

Max Activating Token Index: 51

Text #10

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.8813. Min Act: -0.0002

Data Index: 828178 (C4 (Web Text))

Max Activating Token Index: 859

Text #11

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.8974. Min Act: -0.0002

Data Index: 401500 (C4 (Web Text))

Max Activating Token Index: 40

Text #12

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9088. Min Act: -0.0002

Data Index: 586212 (C4 (Web Text))

Max Activating Token Index: 770

Text #13

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9071. Min Act: -0.0002

Data Index: 84097 (C4 (Web Text))

Max Activating Token Index: 383

Text #14

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9034. Min Act: -0.0002

Data Index: 846539 (C4 (Web Text))

Max Activating Token Index: 617

Text #15

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.8833. Min Act: -0.0002

Data Index: 726034 (C4 (Web Text))

Max Activating Token Index: 51

Text #16

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9002. Min Act: -0.0002

Data Index: 1641850 (Python Code)

Max Activating Token Index: 534

Text #17

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.9068. Min Act: -0.0002

Data Index: 462505 (C4 (Web Text))

Max Activating Token Index: 144

Text #18

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.8830. Min Act: -0.0002

Data Index: 1158677 (C4 (Web Text))

Max Activating Token Index: 916

Text #19

Max Range: 1.0867. Min Range: -1.0867

Max Act: 0.8951. Min Act: -0.0002

Data Index: 736282 (C4 (Web Text))

Max Activating Token Index: 756