Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1038 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.1921. Min Act: -0.1700

Data Index: 503222 (C4 (Web Text))

Max Activating Token Index: 448

Text #1

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0983. Min Act: -0.1700

Data Index: 651738 (C4 (Web Text))

Max Activating Token Index: 1022

Text #2

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.1296. Min Act: -0.1700

Data Index: 381981 (C4 (Web Text))

Max Activating Token Index: 543

Text #3

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0983. Min Act: -0.1700

Data Index: 888811 (C4 (Web Text))

Max Activating Token Index: 38

Text #4

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0358. Min Act: -0.1700

Data Index: 518070 (C4 (Web Text))

Max Activating Token Index: 494

Text #5

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0670. Min Act: -0.1700

Data Index: 20546 (C4 (Web Text))

Max Activating Token Index: 973

Text #6

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0670. Min Act: -0.1700

Data Index: 1106939 (C4 (Web Text))

Max Activating Token Index: 329

Text #7

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0358. Min Act: -0.1700

Data Index: 1117253 (C4 (Web Text))

Max Activating Token Index: 881

Text #8

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0045. Min Act: -0.1700

Data Index: 1211145 (C4 (Web Text))

Max Activating Token Index: 685

Text #9

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9420. Min Act: -0.1700

Data Index: 904863 (C4 (Web Text))

Max Activating Token Index: 754

Text #10

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9732. Min Act: -0.1700

Data Index: 711953 (C4 (Web Text))

Max Activating Token Index: 354

Text #11

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9732. Min Act: -0.1700

Data Index: 337835 (C4 (Web Text))

Max Activating Token Index: 151

Text #12

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9732. Min Act: -0.1700

Data Index: 1137278 (C4 (Web Text))

Max Activating Token Index: 643

Text #13

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9732. Min Act: -0.1700

Data Index: 938800 (C4 (Web Text))

Max Activating Token Index: 710

Text #14

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9732. Min Act: -0.1700

Data Index: 241775 (C4 (Web Text))

Max Activating Token Index: 121

Text #15

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0045. Min Act: -0.1700

Data Index: 300174 (C4 (Web Text))

Max Activating Token Index: 1008

Text #16

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9420. Min Act: -0.1700

Data Index: 43497 (C4 (Web Text))

Max Activating Token Index: 327

Text #17

Max Range: 4.1921. Min Range: -4.1921

Max Act: 4.0045. Min Act: -0.1700

Data Index: 1354254 (C4 (Web Text))

Max Activating Token Index: 241

Text #18

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9420. Min Act: -0.1700

Data Index: 161386 (C4 (Web Text))

Max Activating Token Index: 324

Text #19

Max Range: 4.1921. Min Range: -4.1921

Max Act: 3.9107. Min Act: -0.1700

Data Index: 1562427 (Python Code)

Max Activating Token Index: 24