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

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

Neuron 339 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: 0.5271. Min Range: -0.5271

Max Act: 0.5065. Min Act: -0.0000

Data Index: 1451575 (Python Code)

Max Activating Token Index: 879

Text #1

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.5271. Min Act: -0.0000

Data Index: 485407 (C4 (Web Text))

Max Activating Token Index: 572

Text #2

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4903. Min Act: -0.0000

Data Index: 1640846 (Python Code)

Max Activating Token Index: 186

Text #3

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4937. Min Act: -0.0000

Data Index: 508776 (C4 (Web Text))

Max Activating Token Index: 293

Text #4

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.5151. Min Act: -0.0000

Data Index: 1531675 (Python Code)

Max Activating Token Index: 946

Text #5

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4921. Min Act: -0.0000

Data Index: 1038216 (C4 (Web Text))

Max Activating Token Index: 1001

Text #6

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.5077. Min Act: -0.0000

Data Index: 676630 (C4 (Web Text))

Max Activating Token Index: 414

Text #7

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4782. Min Act: -0.0000

Data Index: 1480851 (Python Code)

Max Activating Token Index: 601

Text #8

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4739. Min Act: -0.0000

Data Index: 215917 (C4 (Web Text))

Max Activating Token Index: 620

Text #9

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4894. Min Act: -0.0000

Data Index: 889979 (C4 (Web Text))

Max Activating Token Index: 174

Text #10

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4217. Min Act: -0.0000

Data Index: 85553 (C4 (Web Text))

Max Activating Token Index: 208

Text #11

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4573. Min Act: -0.0000

Data Index: 1508863 (Python Code)

Max Activating Token Index: 765

Text #12

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4688. Min Act: -0.0000

Data Index: 1198631 (C4 (Web Text))

Max Activating Token Index: 893

Text #13

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4454. Min Act: 0.0000

Data Index: 536618 (C4 (Web Text))

Max Activating Token Index: 560

Text #14

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4282. Min Act: -0.0000

Data Index: 1118738 (C4 (Web Text))

Max Activating Token Index: 744

Text #15

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4411. Min Act: 0.0000

Data Index: 1057427 (C4 (Web Text))

Max Activating Token Index: 478

Text #16

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4642. Min Act: -0.0000

Data Index: 1540548 (Python Code)

Max Activating Token Index: 732

Text #17

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4473. Min Act: -0.0000

Data Index: 74195 (C4 (Web Text))

Max Activating Token Index: 446

Text #18

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4482. Min Act: -0.0000

Data Index: 606358 (C4 (Web Text))

Max Activating Token Index: 912

Text #19

Max Range: 0.5271. Min Range: -0.5271

Max Act: 0.4405. Min Act: -0.0000

Data Index: 940118 (C4 (Web Text))

Max Activating Token Index: 344