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

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

Neuron 500 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1730. Min Act: -0.0000

Data Index: 1441029 (Python Code)

Max Activating Token Index: 705

Text #1

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1966. Min Act: -0.0000

Data Index: 267229 (C4 (Web Text))

Max Activating Token Index: 873

Text #2

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1735. Min Act: -0.0000

Data Index: 1481086 (Python Code)

Max Activating Token Index: 599

Text #3

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1730. Min Act: -0.0000

Data Index: 1352916 (C4 (Web Text))

Max Activating Token Index: 944

Text #4

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1635. Min Act: -0.0000

Data Index: 701516 (C4 (Web Text))

Max Activating Token Index: 20

Text #5

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1685. Min Act: -0.0000

Data Index: 1111623 (C4 (Web Text))

Max Activating Token Index: 1008

Text #6

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1664. Min Act: -0.0000

Data Index: 649443 (C4 (Web Text))

Max Activating Token Index: 661

Text #7

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1589. Min Act: -0.0000

Data Index: 1563738 (Python Code)

Max Activating Token Index: 103

Text #8

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1650. Min Act: -0.0000

Data Index: 336386 (C4 (Web Text))

Max Activating Token Index: 810

Text #9

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1593. Min Act: -0.0000

Data Index: 1285130 (C4 (Web Text))

Max Activating Token Index: 954

Text #10

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1654. Min Act: -0.0000

Data Index: 1582498 (Python Code)

Max Activating Token Index: 24

Text #11

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1592. Min Act: -0.0000

Data Index: 734067 (C4 (Web Text))

Max Activating Token Index: 842

Text #12

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1533. Min Act: -0.0000

Data Index: 1428405 (Python Code)

Max Activating Token Index: 191

Text #13

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1520. Min Act: -0.0000

Data Index: 1094688 (C4 (Web Text))

Max Activating Token Index: 117

Text #14

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1590. Min Act: -0.0000

Data Index: 950116 (C4 (Web Text))

Max Activating Token Index: 1006

Text #15

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1613. Min Act: -0.0000

Data Index: 264348 (C4 (Web Text))

Max Activating Token Index: 708

Text #16

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1558. Min Act: -0.0000

Data Index: 1598249 (Python Code)

Max Activating Token Index: 137

Text #17

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1562. Min Act: -0.0000

Data Index: 1098636 (C4 (Web Text))

Max Activating Token Index: 330

Text #18

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1478. Min Act: -0.0000

Data Index: 266410 (C4 (Web Text))

Max Activating Token Index: 111

Text #19

Max Range: 0.1966. Min Range: -0.1966

Max Act: 0.1476. Min Act: -0.0000

Data Index: 1407153 (Python Code)

Max Activating Token Index: 457