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

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

Neuron 3536 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: 0.0794. Min Range: -0.0794

Max Act: 0.0794. Min Act: -0.0002

Data Index: 968608 (C4 (Web Text))

Max Activating Token Index: 314

Text #1

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0726. Min Act: -0.0002

Data Index: 1595332 (Python Code)

Max Activating Token Index: 568

Text #2

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0662. Min Act: -0.0002

Data Index: 51504 (C4 (Web Text))

Max Activating Token Index: 40

Text #3

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0664. Min Act: -0.0002

Data Index: 677226 (C4 (Web Text))

Max Activating Token Index: 81

Text #4

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0658. Min Act: -0.0002

Data Index: 109193 (C4 (Web Text))

Max Activating Token Index: 757

Text #5

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0626. Min Act: -0.0002

Data Index: 598466 (C4 (Web Text))

Max Activating Token Index: 208

Text #6

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0627. Min Act: -0.0002

Data Index: 839148 (C4 (Web Text))

Max Activating Token Index: 371

Text #7

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0619. Min Act: -0.0002

Data Index: 1573946 (Python Code)

Max Activating Token Index: 44

Text #8

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0611. Min Act: -0.0002

Data Index: 928436 (C4 (Web Text))

Max Activating Token Index: 10

Text #9

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0604. Min Act: -0.0002

Data Index: 795916 (C4 (Web Text))

Max Activating Token Index: 297

Text #10

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0622. Min Act: -0.0002

Data Index: 1529663 (Python Code)

Max Activating Token Index: 406

Text #11

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0614. Min Act: -0.0002

Data Index: 1322557 (C4 (Web Text))

Max Activating Token Index: 277

Text #12

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0597. Min Act: -0.0002

Data Index: 1415459 (Python Code)

Max Activating Token Index: 246

Text #13

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0598. Min Act: -0.0002

Data Index: 111707 (C4 (Web Text))

Max Activating Token Index: 131

Text #14

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0599. Min Act: -0.0002

Data Index: 1495767 (Python Code)

Max Activating Token Index: 18

Text #15

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0598. Min Act: -0.0002

Data Index: 581449 (C4 (Web Text))

Max Activating Token Index: 389

Text #16

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0590. Min Act: -0.0002

Data Index: 1587701 (Python Code)

Max Activating Token Index: 221

Text #17

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0595. Min Act: -0.0002

Data Index: 302395 (C4 (Web Text))

Max Activating Token Index: 668

Text #18

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0598. Min Act: -0.0002

Data Index: 1440259 (Python Code)

Max Activating Token Index: 534

Text #19

Max Range: 0.0794. Min Range: -0.0794

Max Act: 0.0599. Min Act: -0.0002

Data Index: 1438984 (Python Code)

Max Activating Token Index: 270