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

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

Neuron 38 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.1000. Min Act: -0.0001

Data Index: 896559 (C4 (Web Text))

Max Activating Token Index: 885

Text #1

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0863. Min Act: -0.0000

Data Index: 383421 (C4 (Web Text))

Max Activating Token Index: 106

Text #2

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0711. Min Act: -0.0001

Data Index: 1047668 (C4 (Web Text))

Max Activating Token Index: 583

Text #3

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0811. Min Act: -0.0000

Data Index: 774337 (C4 (Web Text))

Max Activating Token Index: 943

Text #4

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0730. Min Act: -0.0001

Data Index: 1427763 (Python Code)

Max Activating Token Index: 302

Text #5

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0674. Min Act: -0.0000

Data Index: 1065330 (C4 (Web Text))

Max Activating Token Index: 491

Text #6

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0646. Min Act: -0.0000

Data Index: 760573 (C4 (Web Text))

Max Activating Token Index: 446

Text #7

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0692. Min Act: -0.0001

Data Index: 726234 (C4 (Web Text))

Max Activating Token Index: 95

Text #8

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0650. Min Act: -0.0001

Data Index: 1080026 (C4 (Web Text))

Max Activating Token Index: 510

Text #9

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0620. Min Act: -0.0000

Data Index: 581547 (C4 (Web Text))

Max Activating Token Index: 774

Text #10

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0556. Min Act: -0.0001

Data Index: 116882 (C4 (Web Text))

Max Activating Token Index: 174

Text #11

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0648. Min Act: -0.0000

Data Index: 377617 (C4 (Web Text))

Max Activating Token Index: 205

Text #12

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0669. Min Act: -0.0001

Data Index: 1575502 (Python Code)

Max Activating Token Index: 639

Text #13

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0622. Min Act: -0.0000

Data Index: 961421 (C4 (Web Text))

Max Activating Token Index: 900

Text #14

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0662. Min Act: -0.0001

Data Index: 1205091 (C4 (Web Text))

Max Activating Token Index: 547

Text #15

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0660. Min Act: -0.0000

Data Index: 867587 (C4 (Web Text))

Max Activating Token Index: 691

Text #16

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0630. Min Act: -0.0000

Data Index: 902414 (C4 (Web Text))

Max Activating Token Index: 583

Text #17

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0624. Min Act: -0.0001

Data Index: 727004 (C4 (Web Text))

Max Activating Token Index: 386

Text #18

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0622. Min Act: -0.0001

Data Index: 1403641 (Python Code)

Max Activating Token Index: 190

Text #19

Max Range: 0.1000. Min Range: -0.1000

Max Act: 0.0595. Min Act: -0.0001

Data Index: 254149 (C4 (Web Text))

Max Activating Token Index: 506