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

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

Neuron 3929 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: 0.8463. Min Range: -0.8463

Max Act: 0.8463. Min Act: -0.0002

Data Index: 132860 (C4 (Web Text))

Max Activating Token Index: 783

Text #1

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.8065. Min Act: -0.0002

Data Index: 225072 (C4 (Web Text))

Max Activating Token Index: 722

Text #2

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.8186. Min Act: -0.0002

Data Index: 1560626 (Python Code)

Max Activating Token Index: 783

Text #3

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.8096. Min Act: -0.0002

Data Index: 1197291 (C4 (Web Text))

Max Activating Token Index: 315

Text #4

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7928. Min Act: -0.0002

Data Index: 295881 (C4 (Web Text))

Max Activating Token Index: 227

Text #5

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7977. Min Act: -0.0002

Data Index: 1527343 (Python Code)

Max Activating Token Index: 781

Text #6

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7753. Min Act: -0.0002

Data Index: 769500 (C4 (Web Text))

Max Activating Token Index: 653

Text #7

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7753. Min Act: -0.0002

Data Index: 1594136 (Python Code)

Max Activating Token Index: 399

Text #8

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7883. Min Act: -0.0002

Data Index: 49336 (C4 (Web Text))

Max Activating Token Index: 806

Text #9

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7722. Min Act: -0.0002

Data Index: 912609 (C4 (Web Text))

Max Activating Token Index: 654

Text #10

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7519. Min Act: -0.0002

Data Index: 723799 (C4 (Web Text))

Max Activating Token Index: 50

Text #11

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7574. Min Act: -0.0002

Data Index: 1325605 (C4 (Web Text))

Max Activating Token Index: 668

Text #12

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7536. Min Act: -0.0002

Data Index: 414733 (C4 (Web Text))

Max Activating Token Index: 35

Text #13

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7596. Min Act: -0.0002

Data Index: 868712 (C4 (Web Text))

Max Activating Token Index: 757

Text #14

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7655. Min Act: -0.0002

Data Index: 1545161 (Python Code)

Max Activating Token Index: 382

Text #15

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7628. Min Act: -0.0002

Data Index: 1328297 (C4 (Web Text))

Max Activating Token Index: 36

Text #16

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7339. Min Act: -0.0002

Data Index: 266838 (C4 (Web Text))

Max Activating Token Index: 23

Text #17

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7347. Min Act: -0.0003

Data Index: 732550 (C4 (Web Text))

Max Activating Token Index: 512

Text #18

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7352. Min Act: -0.0003

Data Index: 1128662 (C4 (Web Text))

Max Activating Token Index: 733

Text #19

Max Range: 0.8463. Min Range: -0.8463

Max Act: 0.7420. Min Act: -0.0003

Data Index: 1225838 (C4 (Web Text))

Max Activating Token Index: 294