Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2723 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0732. Min Act: -0.0001

Data Index: 309970 (C4 (Web Text))

Max Activating Token Index: 360

Text #1

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0726. Min Act: -0.0001

Data Index: 1460265 (Python Code)

Max Activating Token Index: 137

Text #2

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0651. Min Act: -0.0001

Data Index: 89604 (C4 (Web Text))

Max Activating Token Index: 219

Text #3

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0619. Min Act: -0.0001

Data Index: 480673 (C4 (Web Text))

Max Activating Token Index: 214

Text #4

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0644. Min Act: -0.0001

Data Index: 1048467 (C4 (Web Text))

Max Activating Token Index: 175

Text #5

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0577. Min Act: -0.0001

Data Index: 986844 (C4 (Web Text))

Max Activating Token Index: 928

Text #6

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0609. Min Act: -0.0001

Data Index: 467607 (C4 (Web Text))

Max Activating Token Index: 295

Text #7

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0542. Min Act: -0.0001

Data Index: 312646 (C4 (Web Text))

Max Activating Token Index: 47

Text #8

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0560. Min Act: -0.0001

Data Index: 531523 (C4 (Web Text))

Max Activating Token Index: 301

Text #9

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0560. Min Act: -0.0001

Data Index: 1176970 (C4 (Web Text))

Max Activating Token Index: 327

Text #10

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0508. Min Act: -0.0001

Data Index: 1237443 (C4 (Web Text))

Max Activating Token Index: 843

Text #11

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0545. Min Act: -0.0001

Data Index: 1373038 (Python Code)

Max Activating Token Index: 182

Text #12

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0523. Min Act: -0.0001

Data Index: 357781 (C4 (Web Text))

Max Activating Token Index: 722

Text #13

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0542. Min Act: -0.0001

Data Index: 1047021 (C4 (Web Text))

Max Activating Token Index: 325

Text #14

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0544. Min Act: -0.0001

Data Index: 214914 (C4 (Web Text))

Max Activating Token Index: 807

Text #15

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0582. Min Act: -0.0001

Data Index: 587230 (C4 (Web Text))

Max Activating Token Index: 693

Text #16

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0557. Min Act: -0.0001

Data Index: 223088 (C4 (Web Text))

Max Activating Token Index: 907

Text #17

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0481. Min Act: -0.0001

Data Index: 1353660 (C4 (Web Text))

Max Activating Token Index: 156

Text #18

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0531. Min Act: -0.0001

Data Index: 514758 (C4 (Web Text))

Max Activating Token Index: 445

Text #19

Max Range: 0.0732. Min Range: -0.0732

Max Act: 0.0493. Min Act: -0.0001

Data Index: 430443 (C4 (Web Text))

Max Activating Token Index: 192