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

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

Neuron 2643 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.7688. Min Act: -0.0000

Data Index: 945914 (C4 (Web Text))

Max Activating Token Index: 452

Text #1

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.6900. Min Act: -0.0000

Data Index: 759454 (C4 (Web Text))

Max Activating Token Index: 662

Text #2

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.6668. Min Act: -0.0000

Data Index: 453941 (C4 (Web Text))

Max Activating Token Index: 301

Text #3

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.5900. Min Act: -0.0000

Data Index: 1571055 (Python Code)

Max Activating Token Index: 273

Text #4

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.5452. Min Act: -0.0000

Data Index: 538219 (C4 (Web Text))

Max Activating Token Index: 869

Text #5

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.5609. Min Act: -0.0000

Data Index: 1510923 (Python Code)

Max Activating Token Index: 291

Text #6

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.5575. Min Act: -0.0000

Data Index: 941273 (C4 (Web Text))

Max Activating Token Index: 1014

Text #7

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.5624. Min Act: -0.0000

Data Index: 848256 (C4 (Web Text))

Max Activating Token Index: 720

Text #8

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.4673. Min Act: -0.0000

Data Index: 1607036 (Python Code)

Max Activating Token Index: 216

Text #9

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3829. Min Act: -0.0000

Data Index: 1396806 (Python Code)

Max Activating Token Index: 88

Text #10

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.4183. Min Act: -0.0000

Data Index: 122008 (C4 (Web Text))

Max Activating Token Index: 262

Text #11

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.4348. Min Act: -0.0000

Data Index: 1527879 (Python Code)

Max Activating Token Index: 667

Text #12

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.4332. Min Act: -0.0000

Data Index: 1360622 (Python Code)

Max Activating Token Index: 474

Text #13

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.4085. Min Act: -0.0000

Data Index: 1475241 (Python Code)

Max Activating Token Index: 63

Text #14

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3703. Min Act: -0.0000

Data Index: 1512676 (Python Code)

Max Activating Token Index: 473

Text #15

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3622. Min Act: -0.0000

Data Index: 1443810 (Python Code)

Max Activating Token Index: 778

Text #16

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3880. Min Act: 0.0000

Data Index: 1480426 (Python Code)

Max Activating Token Index: 95

Text #17

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3877. Min Act: -0.0000

Data Index: 1489407 (Python Code)

Max Activating Token Index: 901

Text #18

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3525. Min Act: -0.0000

Data Index: 1374587 (Python Code)

Max Activating Token Index: 930

Text #19

Max Range: 0.7688. Min Range: -0.7688

Max Act: 0.3678. Min Act: -0.0000

Data Index: 1549191 (Python Code)

Max Activating Token Index: 862