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

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

Neuron 2756 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: 4.6903. Min Range: -4.6903

Max Act: 4.5776. Min Act: -0.0000

Data Index: 727185 (C4 (Web Text))

Max Activating Token Index: 111

Text #1

Max Range: 4.6903. Min Range: -4.6903

Max Act: 4.6903. Min Act: -0.0000

Data Index: 1483159 (Python Code)

Max Activating Token Index: 856

Text #2

Max Range: 4.6903. Min Range: -4.6903

Max Act: 4.6349. Min Act: -0.0000

Data Index: 1500184 (Python Code)

Max Activating Token Index: 928

Text #3

Max Range: 4.6903. Min Range: -4.6903

Max Act: 4.6417. Min Act: -0.0000

Data Index: 1609595 (Python Code)

Max Activating Token Index: 79

Text #4

Max Range: 4.6903. Min Range: -4.6903

Max Act: 4.0345. Min Act: -0.0000

Data Index: 1586310 (Python Code)

Max Activating Token Index: 20

Text #5

Max Range: 4.6903. Min Range: -4.6903

Max Act: 4.0278. Min Act: -0.0000

Data Index: 1562297 (Python Code)

Max Activating Token Index: 794

Text #6

Max Range: 4.6903. Min Range: -4.6903

Max Act: 4.2123. Min Act: -0.0000

Data Index: 239453 (C4 (Web Text))

Max Activating Token Index: 1013

Text #7

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.8377. Min Act: -0.0000

Data Index: 1595973 (Python Code)

Max Activating Token Index: 28

Text #8

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.8148. Min Act: -0.0000

Data Index: 1626798 (Python Code)

Max Activating Token Index: 19

Text #9

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.7449. Min Act: -0.0000

Data Index: 369917 (C4 (Web Text))

Max Activating Token Index: 87

Text #10

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.8963. Min Act: -0.0000

Data Index: 999764 (C4 (Web Text))

Max Activating Token Index: 603

Text #11

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.7558. Min Act: -0.0000

Data Index: 1524891 (Python Code)

Max Activating Token Index: 3

Text #12

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.7635. Min Act: -0.0000

Data Index: 1641019 (Python Code)

Max Activating Token Index: 262

Text #13

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.6949. Min Act: -0.0000

Data Index: 1383557 (Python Code)

Max Activating Token Index: 60

Text #14

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.5828. Min Act: -0.0000

Data Index: 254523 (C4 (Web Text))

Max Activating Token Index: 644

Text #15

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.1484. Min Act: -0.0000

Data Index: 1506227 (Python Code)

Max Activating Token Index: 393

Text #16

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.4235. Min Act: -0.0000

Data Index: 718998 (C4 (Web Text))

Max Activating Token Index: 547

Text #17

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.2665. Min Act: -0.0000

Data Index: 1588178 (Python Code)

Max Activating Token Index: 137

Text #18

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.4138. Min Act: -0.0000

Data Index: 421050 (C4 (Web Text))

Max Activating Token Index: 583

Text #19

Max Range: 4.6903. Min Range: -4.6903

Max Act: 3.0637. Min Act: -0.0000

Data Index: 144630 (C4 (Web Text))

Max Activating Token Index: 120