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

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

Neuron 1829 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: 0.4291. Min Range: -0.4291

Max Act: 0.4081. Min Act: -0.0000

Data Index: 1065245 (C4 (Web Text))

Max Activating Token Index: 836

Text #1

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.4139. Min Act: -0.0000

Data Index: 214110 (C4 (Web Text))

Max Activating Token Index: 654

Text #2

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3863. Min Act: -0.0000

Data Index: 1014090 (C4 (Web Text))

Max Activating Token Index: 338

Text #3

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.4150. Min Act: -0.0000

Data Index: 872144 (C4 (Web Text))

Max Activating Token Index: 1002

Text #4

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3839. Min Act: -0.0000

Data Index: 1641085 (Python Code)

Max Activating Token Index: 962

Text #5

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.4074. Min Act: -0.0000

Data Index: 280012 (C4 (Web Text))

Max Activating Token Index: 883

Text #6

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3902. Min Act: -0.0000

Data Index: 1444991 (Python Code)

Max Activating Token Index: 17

Text #7

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3986. Min Act: -0.0000

Data Index: 133579 (C4 (Web Text))

Max Activating Token Index: 662

Text #8

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3622. Min Act: -0.0000

Data Index: 1454400 (Python Code)

Max Activating Token Index: 689

Text #9

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.4291. Min Act: -0.0000

Data Index: 637292 (C4 (Web Text))

Max Activating Token Index: 509

Text #10

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3812. Min Act: -0.0000

Data Index: 307875 (C4 (Web Text))

Max Activating Token Index: 1015

Text #11

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3846. Min Act: -0.0000

Data Index: 918832 (C4 (Web Text))

Max Activating Token Index: 341

Text #12

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.4008. Min Act: -0.0000

Data Index: 620924 (C4 (Web Text))

Max Activating Token Index: 875

Text #13

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3670. Min Act: -0.0000

Data Index: 670045 (C4 (Web Text))

Max Activating Token Index: 517

Text #14

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3653. Min Act: -0.0000

Data Index: 466600 (C4 (Web Text))

Max Activating Token Index: 838

Text #15

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3681. Min Act: -0.0000

Data Index: 1611147 (Python Code)

Max Activating Token Index: 151

Text #16

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3564. Min Act: -0.0000

Data Index: 487921 (C4 (Web Text))

Max Activating Token Index: 635

Text #17

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3655. Min Act: -0.0000

Data Index: 1461718 (Python Code)

Max Activating Token Index: 189

Text #18

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3637. Min Act: -0.0000

Data Index: 486247 (C4 (Web Text))

Max Activating Token Index: 706

Text #19

Max Range: 0.4291. Min Range: -0.4291

Max Act: 0.3486. Min Act: -0.0000

Data Index: 127159 (C4 (Web Text))

Max Activating Token Index: 597