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

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

Neuron 352 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.8113. Min Act: -0.1700

Data Index: 939890 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.7801. Min Act: -0.1700

Data Index: 584942 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.7488. Min Act: -0.1700

Data Index: 1347968 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6862. Min Act: -0.1700

Data Index: 811794 (C4 (Web Text))

Max Activating Token Index: 440

Text #4

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6862. Min Act: -0.1700

Data Index: 596767 (C4 (Web Text))

Max Activating Token Index: 199

Text #5

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6236. Min Act: -0.1700

Data Index: 251990 (C4 (Web Text))

Max Activating Token Index: 886

Text #6

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6236. Min Act: -0.1700

Data Index: 504695 (C4 (Web Text))

Max Activating Token Index: 848

Text #7

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6236. Min Act: -0.1700

Data Index: 1545604 (Python Code)

Max Activating Token Index: 850

Text #8

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6236. Min Act: -0.1700

Data Index: 768134 (C4 (Web Text))

Max Activating Token Index: 902

Text #9

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6549. Min Act: -0.1700

Data Index: 330315 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5609. Min Act: -0.1700

Data Index: 1132828 (C4 (Web Text))

Max Activating Token Index: 262

Text #11

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5923. Min Act: -0.1700

Data Index: 405007 (C4 (Web Text))

Max Activating Token Index: 902

Text #12

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6549. Min Act: -0.1700

Data Index: 333639 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6236. Min Act: -0.1700

Data Index: 336851 (C4 (Web Text))

Max Activating Token Index: 783

Text #14

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5609. Min Act: -0.1700

Data Index: 151119 (C4 (Web Text))

Max Activating Token Index: 233

Text #15

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.6236. Min Act: -0.1700

Data Index: 213910 (C4 (Web Text))

Max Activating Token Index: 838

Text #16

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5609. Min Act: -0.1700

Data Index: 65154 (C4 (Web Text))

Max Activating Token Index: 824

Text #17

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5609. Min Act: -0.1700

Data Index: 314003 (C4 (Web Text))

Max Activating Token Index: 894

Text #18

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5923. Min Act: -0.1700

Data Index: 820516 (C4 (Web Text))

Max Activating Token Index: 396

Text #19

Max Range: 3.8113. Min Range: -3.8113

Max Act: 3.5296. Min Act: -0.1700

Data Index: 138384 (C4 (Web Text))

Max Activating Token Index: 788