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

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

Neuron 970 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.8819. Min Act: -0.0001

Data Index: 1608942 (Python Code)

Max Activating Token Index: 304

Text #1

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7779. Min Act: -0.0001

Data Index: 53700 (C4 (Web Text))

Max Activating Token Index: 173

Text #2

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.8144. Min Act: -0.0001

Data Index: 967865 (C4 (Web Text))

Max Activating Token Index: 567

Text #3

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.8573. Min Act: -0.0001

Data Index: 121491 (C4 (Web Text))

Max Activating Token Index: 287

Text #4

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7976. Min Act: -0.0001

Data Index: 26888 (C4 (Web Text))

Max Activating Token Index: 995

Text #5

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7420. Min Act: -0.0001

Data Index: 415847 (C4 (Web Text))

Max Activating Token Index: 252

Text #6

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7519. Min Act: -0.0001

Data Index: 444655 (C4 (Web Text))

Max Activating Token Index: 738

Text #7

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7680. Min Act: -0.0001

Data Index: 496312 (C4 (Web Text))

Max Activating Token Index: 404

Text #8

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7641. Min Act: -0.0001

Data Index: 282548 (C4 (Web Text))

Max Activating Token Index: 257

Text #9

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7398. Min Act: -0.0001

Data Index: 1234255 (C4 (Web Text))

Max Activating Token Index: 920

Text #10

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7061. Min Act: -0.0001

Data Index: 356062 (C4 (Web Text))

Max Activating Token Index: 1005

Text #11

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7660. Min Act: -0.0001

Data Index: 344659 (C4 (Web Text))

Max Activating Token Index: 940

Text #12

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7412. Min Act: -0.0001

Data Index: 631367 (C4 (Web Text))

Max Activating Token Index: 298

Text #13

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7196. Min Act: -0.0001

Data Index: 576814 (C4 (Web Text))

Max Activating Token Index: 883

Text #14

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7387. Min Act: -0.0001

Data Index: 1590364 (Python Code)

Max Activating Token Index: 863

Text #15

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7593. Min Act: -0.0001

Data Index: 94833 (C4 (Web Text))

Max Activating Token Index: 479

Text #16

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7338. Min Act: -0.0001

Data Index: 22302 (C4 (Web Text))

Max Activating Token Index: 775

Text #17

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.6421. Min Act: -0.0001

Data Index: 368003 (C4 (Web Text))

Max Activating Token Index: 553

Text #18

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7514. Min Act: -0.0001

Data Index: 359725 (C4 (Web Text))

Max Activating Token Index: 632

Text #19

Max Range: 0.8819. Min Range: -0.8819

Max Act: 0.7294. Min Act: -0.0001

Data Index: 755325 (C4 (Web Text))

Max Activating Token Index: 634