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

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

Neuron 1533 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.2418. Min Act: -0.1700

Data Index: 1031601 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.2261. Min Act: -0.1700

Data Index: 644268 (C4 (Web Text))

Max Activating Token Index: 10

Text #2

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.2261. Min Act: -0.1700

Data Index: 215966 (C4 (Web Text))

Max Activating Token Index: 625

Text #3

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1631. Min Act: -0.1700

Data Index: 1189729 (C4 (Web Text))

Max Activating Token Index: 64

Text #4

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1474. Min Act: -0.1700

Data Index: 872569 (C4 (Web Text))

Max Activating Token Index: 520

Text #5

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1316. Min Act: -0.1700

Data Index: 985680 (C4 (Web Text))

Max Activating Token Index: 86

Text #6

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1474. Min Act: -0.1700

Data Index: 551480 (C4 (Web Text))

Max Activating Token Index: 626

Text #7

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1001. Min Act: -0.1700

Data Index: 1326162 (C4 (Web Text))

Max Activating Token Index: 27

Text #8

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0686. Min Act: -0.1700

Data Index: 681257 (C4 (Web Text))

Max Activating Token Index: 935

Text #9

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1001. Min Act: -0.1700

Data Index: 1357536 (C4 (Web Text))

Max Activating Token Index: 479

Text #10

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.1159. Min Act: -0.1700

Data Index: 887535 (C4 (Web Text))

Max Activating Token Index: 30

Text #11

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0528. Min Act: -0.1700

Data Index: 726519 (C4 (Web Text))

Max Activating Token Index: 177

Text #12

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0843. Min Act: -0.1700

Data Index: 964810 (C4 (Web Text))

Max Activating Token Index: 32

Text #13

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0843. Min Act: -0.1700

Data Index: 756593 (C4 (Web Text))

Max Activating Token Index: 246

Text #14

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0370. Min Act: -0.1700

Data Index: 1098700 (C4 (Web Text))

Max Activating Token Index: 54

Text #15

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0370. Min Act: -0.1700

Data Index: 971372 (C4 (Web Text))

Max Activating Token Index: 701

Text #16

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0370. Min Act: -0.1700

Data Index: 720334 (C4 (Web Text))

Max Activating Token Index: 265

Text #17

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0370. Min Act: -0.1700

Data Index: 569072 (C4 (Web Text))

Max Activating Token Index: 72

Text #18

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0528. Min Act: -0.1700

Data Index: 198902 (C4 (Web Text))

Max Activating Token Index: 584

Text #19

Max Range: 3.2418. Min Range: -3.2418

Max Act: 3.0370. Min Act: -0.1700

Data Index: 1535088 (Python Code)

Max Activating Token Index: 24