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

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

Neuron 2624 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.4660. Min Act: -0.0000

Data Index: 1364300 (Python Code)

Max Activating Token Index: 896

Text #1

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.4119. Min Act: -0.0000

Data Index: 919804 (C4 (Web Text))

Max Activating Token Index: 946

Text #2

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.1427. Min Act: -0.0000

Data Index: 1589100 (Python Code)

Max Activating Token Index: 210

Text #3

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.3417. Min Act: -0.0000

Data Index: 1113536 (C4 (Web Text))

Max Activating Token Index: 545

Text #4

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.4728. Min Act: -0.0000

Data Index: 1300381 (C4 (Web Text))

Max Activating Token Index: 228

Text #5

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.0625. Min Act: -0.0000

Data Index: 395413 (C4 (Web Text))

Max Activating Token Index: 1001

Text #6

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.3686. Min Act: -0.0000

Data Index: 1647630 (Python Code)

Max Activating Token Index: 834

Text #7

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9728. Min Act: -0.0000

Data Index: 1272597 (C4 (Web Text))

Max Activating Token Index: 671

Text #8

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.0762. Min Act: -0.0000

Data Index: 1245901 (C4 (Web Text))

Max Activating Token Index: 311

Text #9

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.0710. Min Act: -0.0000

Data Index: 102095 (C4 (Web Text))

Max Activating Token Index: 766

Text #10

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.0277. Min Act: -0.0000

Data Index: 1413678 (Python Code)

Max Activating Token Index: 660

Text #11

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.1800. Min Act: -0.0000

Data Index: 1215412 (C4 (Web Text))

Max Activating Token Index: 252

Text #12

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.8727. Min Act: -0.0000

Data Index: 915385 (C4 (Web Text))

Max Activating Token Index: 663

Text #13

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9996. Min Act: -0.0000

Data Index: 1412056 (Python Code)

Max Activating Token Index: 894

Text #14

Max Range: 3.4728. Min Range: -3.4728

Max Act: 3.1246. Min Act: -0.0000

Data Index: 1332925 (C4 (Web Text))

Max Activating Token Index: 477

Text #15

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9699. Min Act: -0.0000

Data Index: 1553050 (Python Code)

Max Activating Token Index: 338

Text #16

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9597. Min Act: -0.0000

Data Index: 174699 (C4 (Web Text))

Max Activating Token Index: 506

Text #17

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9631. Min Act: -0.0000

Data Index: 1416615 (Python Code)

Max Activating Token Index: 572

Text #18

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9511. Min Act: -0.0000

Data Index: 1624892 (Python Code)

Max Activating Token Index: 114

Text #19

Max Range: 3.4728. Min Range: -3.4728

Max Act: 2.9214. Min Act: -0.0000

Data Index: 171669 (C4 (Web Text))

Max Activating Token Index: 625