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

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

Neuron 1958 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: 3.1324. Min Range: -3.1324

Max Act: 3.1324. Min Act: -0.1700

Data Index: 1402440 (Python Code)

Max Activating Token Index: 8

Text #1

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0851. Min Act: -0.1700

Data Index: 688044 (C4 (Web Text))

Max Activating Token Index: 18

Text #2

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0536. Min Act: -0.1700

Data Index: 1309667 (C4 (Web Text))

Max Activating Token Index: 2

Text #3

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0536. Min Act: -0.1700

Data Index: 1614792 (Python Code)

Max Activating Token Index: 44

Text #4

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0220. Min Act: -0.1700

Data Index: 1620920 (Python Code)

Max Activating Token Index: 814

Text #5

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0378. Min Act: -0.1700

Data Index: 1453570 (Python Code)

Max Activating Token Index: 815

Text #6

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0220. Min Act: -0.1700

Data Index: 1518191 (Python Code)

Max Activating Token Index: 39

Text #7

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0536. Min Act: -0.1700

Data Index: 1641838 (Python Code)

Max Activating Token Index: 2

Text #8

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0220. Min Act: -0.1700

Data Index: 1444184 (Python Code)

Max Activating Token Index: 873

Text #9

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0220. Min Act: -0.1700

Data Index: 1176273 (C4 (Web Text))

Max Activating Token Index: 3

Text #10

Max Range: 3.1324. Min Range: -3.1324

Max Act: 3.0062. Min Act: -0.1700

Data Index: 1650034 (Python Code)

Max Activating Token Index: 4

Text #11

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9904. Min Act: -0.1700

Data Index: 1329358 (C4 (Web Text))

Max Activating Token Index: 34

Text #12

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9904. Min Act: -0.1700

Data Index: 1452501 (Python Code)

Max Activating Token Index: 885

Text #13

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9904. Min Act: -0.1700

Data Index: 647793 (C4 (Web Text))

Max Activating Token Index: 9

Text #14

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9904. Min Act: -0.1700

Data Index: 670669 (C4 (Web Text))

Max Activating Token Index: 5

Text #15

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9904. Min Act: -0.1700

Data Index: 1555815 (Python Code)

Max Activating Token Index: 691

Text #16

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9587. Min Act: -0.1700

Data Index: 1513996 (Python Code)

Max Activating Token Index: 576

Text #17

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9587. Min Act: -0.1700

Data Index: 1552918 (Python Code)

Max Activating Token Index: 40

Text #18

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9904. Min Act: -0.1700

Data Index: 1371589 (Python Code)

Max Activating Token Index: 16

Text #19

Max Range: 3.1324. Min Range: -3.1324

Max Act: 2.9587. Min Act: -0.1700

Data Index: 1511430 (Python Code)

Max Activating Token Index: 808