Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 733 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.8738. Min Act: -0.1700

Data Index: 931011 (C4 (Web Text))

Max Activating Token Index: 190

Text #1

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.8738. Min Act: -0.1700

Data Index: 75865 (C4 (Web Text))

Max Activating Token Index: 260

Text #2

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.8269. Min Act: -0.1700

Data Index: 1383948 (Python Code)

Max Activating Token Index: 6

Text #3

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.8113. Min Act: -0.1700

Data Index: 1531673 (Python Code)

Max Activating Token Index: 38

Text #4

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7800. Min Act: -0.1700

Data Index: 1501231 (Python Code)

Max Activating Token Index: 13

Text #5

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7643. Min Act: -0.1700

Data Index: 1410716 (Python Code)

Max Activating Token Index: 30

Text #6

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7800. Min Act: -0.1700

Data Index: 135323 (C4 (Web Text))

Max Activating Token Index: 556

Text #7

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7487. Min Act: -0.1700

Data Index: 1460597 (Python Code)

Max Activating Token Index: 18

Text #8

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7487. Min Act: -0.1700

Data Index: 773895 (C4 (Web Text))

Max Activating Token Index: 444

Text #9

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7331. Min Act: -0.1700

Data Index: 1522298 (Python Code)

Max Activating Token Index: 51

Text #10

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7018. Min Act: -0.1700

Data Index: 190687 (C4 (Web Text))

Max Activating Token Index: 647

Text #11

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7018. Min Act: -0.1700

Data Index: 1458356 (Python Code)

Max Activating Token Index: 462

Text #12

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7018. Min Act: -0.1700

Data Index: 407677 (C4 (Web Text))

Max Activating Token Index: 466

Text #13

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7018. Min Act: -0.1700

Data Index: 1639251 (Python Code)

Max Activating Token Index: 501

Text #14

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.7018. Min Act: -0.1700

Data Index: 1651417 (Python Code)

Max Activating Token Index: 34

Text #15

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.6861. Min Act: -0.1700

Data Index: 1428675 (Python Code)

Max Activating Token Index: 14

Text #16

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.6861. Min Act: -0.1700

Data Index: 1558484 (Python Code)

Max Activating Token Index: 599

Text #17

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.6705. Min Act: -0.1700

Data Index: 1437376 (Python Code)

Max Activating Token Index: 927

Text #18

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.6548. Min Act: -0.1700

Data Index: 770172 (C4 (Web Text))

Max Activating Token Index: 6

Text #19

Max Range: 3.8738. Min Range: -3.8738

Max Act: 3.6861. Min Act: -0.1700

Data Index: 1044993 (C4 (Web Text))

Max Activating Token Index: 119