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

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

Neuron 1181 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1014. Min Range: -5.1014

Max Act: 5.1014. Min Act: -0.1700

Data Index: 1503532 (Python Code)

Max Activating Token Index: 648

Text #1

Max Range: 5.1014. Min Range: -5.1014

Max Act: 5.0076. Min Act: -0.1700

Data Index: 430001 (C4 (Web Text))

Max Activating Token Index: 914

Text #2

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8514. Min Act: -0.1700

Data Index: 379940 (C4 (Web Text))

Max Activating Token Index: 678

Text #3

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9139. Min Act: -0.1700

Data Index: 1021229 (C4 (Web Text))

Max Activating Token Index: 874

Text #4

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9139. Min Act: -0.1700

Data Index: 94202 (C4 (Web Text))

Max Activating Token Index: 1013

Text #5

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9764. Min Act: -0.1700

Data Index: 1604632 (Python Code)

Max Activating Token Index: 326

Text #6

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8826. Min Act: -0.1700

Data Index: 685479 (C4 (Web Text))

Max Activating Token Index: 919

Text #7

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9139. Min Act: -0.1700

Data Index: 420846 (C4 (Web Text))

Max Activating Token Index: 777

Text #8

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8514. Min Act: -0.1700

Data Index: 1441561 (Python Code)

Max Activating Token Index: 210

Text #9

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9764. Min Act: -0.1700

Data Index: 1502885 (Python Code)

Max Activating Token Index: 528

Text #10

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9139. Min Act: -0.1700

Data Index: 1286937 (C4 (Web Text))

Max Activating Token Index: 811

Text #11

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9451. Min Act: -0.1700

Data Index: 1054838 (C4 (Web Text))

Max Activating Token Index: 352

Text #12

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9139. Min Act: -0.1700

Data Index: 852668 (C4 (Web Text))

Max Activating Token Index: 822

Text #13

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.9139. Min Act: -0.1700

Data Index: 840538 (C4 (Web Text))

Max Activating Token Index: 687

Text #14

Max Range: 5.1014. Min Range: -5.1014

Max Act: 5.0076. Min Act: -0.1700

Data Index: 699667 (C4 (Web Text))

Max Activating Token Index: 841

Text #15

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8826. Min Act: -0.1700

Data Index: 82938 (C4 (Web Text))

Max Activating Token Index: 1008

Text #16

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8514. Min Act: -0.1700

Data Index: 1368406 (Python Code)

Max Activating Token Index: 259

Text #17

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8826. Min Act: -0.1700

Data Index: 609711 (C4 (Web Text))

Max Activating Token Index: 862

Text #18

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8826. Min Act: -0.1700

Data Index: 135127 (C4 (Web Text))

Max Activating Token Index: 342

Text #19

Max Range: 5.1014. Min Range: -5.1014

Max Act: 4.8826. Min Act: -0.1700

Data Index: 862240 (C4 (Web Text))

Max Activating Token Index: 1001