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

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

Neuron 2705 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3098. Min Range: -1.3098

Max Act: 1.3098. Min Act: 0.0000

Data Index: 1534811 (Python Code)

Max Activating Token Index: 80

Text #1

Max Range: 1.3098. Min Range: -1.3098

Max Act: 1.0367. Min Act: -0.0000

Data Index: 90090 (C4 (Web Text))

Max Activating Token Index: 535

Text #2

Max Range: 1.3098. Min Range: -1.3098

Max Act: 1.0491. Min Act: -0.0000

Data Index: 1147197 (C4 (Web Text))

Max Activating Token Index: 550

Text #3

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8951. Min Act: -0.0000

Data Index: 1640082 (Python Code)

Max Activating Token Index: 1014

Text #4

Max Range: 1.3098. Min Range: -1.3098

Max Act: 1.0467. Min Act: -0.0000

Data Index: 1299531 (C4 (Web Text))

Max Activating Token Index: 514

Text #5

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.7937. Min Act: -0.0000

Data Index: 1369631 (Python Code)

Max Activating Token Index: 280

Text #6

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8941. Min Act: -0.0000

Data Index: 1593955 (Python Code)

Max Activating Token Index: 331

Text #7

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8748. Min Act: -0.0000

Data Index: 339124 (C4 (Web Text))

Max Activating Token Index: 21

Text #8

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8932. Min Act: -0.0000

Data Index: 1589599 (Python Code)

Max Activating Token Index: 651

Text #9

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.9514. Min Act: -0.0000

Data Index: 1469024 (Python Code)

Max Activating Token Index: 267

Text #10

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8903. Min Act: -0.0000

Data Index: 1173452 (C4 (Web Text))

Max Activating Token Index: 40

Text #11

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8640. Min Act: -0.0000

Data Index: 720746 (C4 (Web Text))

Max Activating Token Index: 918

Text #12

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8609. Min Act: -0.0000

Data Index: 507747 (C4 (Web Text))

Max Activating Token Index: 845

Text #13

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8852. Min Act: 0.0000

Data Index: 1573100 (Python Code)

Max Activating Token Index: 401

Text #14

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8524. Min Act: -0.0000

Data Index: 1403083 (Python Code)

Max Activating Token Index: 614

Text #15

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8802. Min Act: -0.0000

Data Index: 975527 (C4 (Web Text))

Max Activating Token Index: 936

Text #16

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8224. Min Act: -0.0000

Data Index: 1384831 (Python Code)

Max Activating Token Index: 55

Text #17

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.8493. Min Act: -0.0000

Data Index: 1109826 (C4 (Web Text))

Max Activating Token Index: 579

Text #18

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.7965. Min Act: -0.0000

Data Index: 1641288 (Python Code)

Max Activating Token Index: 506

Text #19

Max Range: 1.3098. Min Range: -1.3098

Max Act: 0.7943. Min Act: 0.0000

Data Index: 1616098 (Python Code)

Max Activating Token Index: 153