Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 814 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0765. Min Act: -0.0003

Data Index: 345938 (C4 (Web Text))

Max Activating Token Index: 35

Text #1

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0719. Min Act: -0.0003

Data Index: 768117 (C4 (Web Text))

Max Activating Token Index: 308

Text #2

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0601. Min Act: -0.0004

Data Index: 466293 (C4 (Web Text))

Max Activating Token Index: 33

Text #3

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0605. Min Act: -0.0003

Data Index: 22419 (C4 (Web Text))

Max Activating Token Index: 40

Text #4

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0473. Min Act: -0.0003

Data Index: 709759 (C4 (Web Text))

Max Activating Token Index: 640

Text #5

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0579. Min Act: -0.0003

Data Index: 636342 (C4 (Web Text))

Max Activating Token Index: 507

Text #6

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0545. Min Act: -0.0003

Data Index: 1049604 (C4 (Web Text))

Max Activating Token Index: 656

Text #7

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0516. Min Act: -0.0003

Data Index: 299051 (C4 (Web Text))

Max Activating Token Index: 1006

Text #8

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0499. Min Act: -0.0003

Data Index: 470079 (C4 (Web Text))

Max Activating Token Index: 130

Text #9

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0489. Min Act: -0.0003

Data Index: 899870 (C4 (Web Text))

Max Activating Token Index: 543

Text #10

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0498. Min Act: -0.0003

Data Index: 666264 (C4 (Web Text))

Max Activating Token Index: 560

Text #11

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0521. Min Act: -0.0003

Data Index: 279734 (C4 (Web Text))

Max Activating Token Index: 333

Text #12

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0488. Min Act: -0.0003

Data Index: 324580 (C4 (Web Text))

Max Activating Token Index: 103

Text #13

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0446. Min Act: -0.0003

Data Index: 1165242 (C4 (Web Text))

Max Activating Token Index: 652

Text #14

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0506. Min Act: -0.0003

Data Index: 688800 (C4 (Web Text))

Max Activating Token Index: 153

Text #15

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0493. Min Act: -0.0003

Data Index: 337663 (C4 (Web Text))

Max Activating Token Index: 11

Text #16

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0392. Min Act: -0.0004

Data Index: 37435 (C4 (Web Text))

Max Activating Token Index: 651

Text #17

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0440. Min Act: -0.0003

Data Index: 1113815 (C4 (Web Text))

Max Activating Token Index: 124

Text #18

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0458. Min Act: -0.0003

Data Index: 1165886 (C4 (Web Text))

Max Activating Token Index: 464

Text #19

Max Range: 0.0765. Min Range: -0.0765

Max Act: 0.0440. Min Act: -0.0003

Data Index: 665517 (C4 (Web Text))

Max Activating Token Index: 277