Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1637 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1995. Min Act: -0.0001

Data Index: 1633623 (Python Code)

Max Activating Token Index: 862

Text #1

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1926. Min Act: -0.0001

Data Index: 1511472 (Python Code)

Max Activating Token Index: 921

Text #2

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1657. Min Act: -0.0000

Data Index: 682465 (C4 (Web Text))

Max Activating Token Index: 761

Text #3

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1623. Min Act: -0.0001

Data Index: 1193072 (C4 (Web Text))

Max Activating Token Index: 566

Text #4

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1774. Min Act: -0.0001

Data Index: 217224 (C4 (Web Text))

Max Activating Token Index: 563

Text #5

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1606. Min Act: -0.0001

Data Index: 1459869 (Python Code)

Max Activating Token Index: 42

Text #6

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1649. Min Act: -0.0000

Data Index: 989794 (C4 (Web Text))

Max Activating Token Index: 695

Text #7

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1611. Min Act: -0.0001

Data Index: 1425640 (Python Code)

Max Activating Token Index: 414

Text #8

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1455. Min Act: -0.0001

Data Index: 1387778 (Python Code)

Max Activating Token Index: 277

Text #9

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1571. Min Act: -0.0001

Data Index: 1208491 (C4 (Web Text))

Max Activating Token Index: 855

Text #10

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1510. Min Act: -0.0001

Data Index: 670880 (C4 (Web Text))

Max Activating Token Index: 159

Text #11

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1499. Min Act: -0.0001

Data Index: 1628370 (Python Code)

Max Activating Token Index: 170

Text #12

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1478. Min Act: -0.0000

Data Index: 1020142 (C4 (Web Text))

Max Activating Token Index: 936

Text #13

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1479. Min Act: -0.0000

Data Index: 113403 (C4 (Web Text))

Max Activating Token Index: 618

Text #14

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1531. Min Act: -0.0001

Data Index: 688351 (C4 (Web Text))

Max Activating Token Index: 637

Text #15

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1572. Min Act: -0.0001

Data Index: 1509575 (Python Code)

Max Activating Token Index: 310

Text #16

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1475. Min Act: -0.0000

Data Index: 936921 (C4 (Web Text))

Max Activating Token Index: 391

Text #17

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1406. Min Act: -0.0000

Data Index: 571241 (C4 (Web Text))

Max Activating Token Index: 400

Text #18

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1457. Min Act: -0.0000

Data Index: 1552026 (Python Code)

Max Activating Token Index: 154

Text #19

Max Range: 0.1995. Min Range: -0.1995

Max Act: 0.1511. Min Act: -0.0000

Data Index: 857638 (C4 (Web Text))

Max Activating Token Index: 24