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

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

Neuron 1575 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: 5.9830. Min Range: -5.9830

Max Act: 5.9518. Min Act: -0.1700

Data Index: 341535 (C4 (Web Text))

Max Activating Token Index: 909

Text #1

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.9830. Min Act: -0.1700

Data Index: 98562 (C4 (Web Text))

Max Activating Token Index: 834

Text #2

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7955. Min Act: -0.1700

Data Index: 398001 (C4 (Web Text))

Max Activating Token Index: 967

Text #3

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.8580. Min Act: -0.1700

Data Index: 357990 (C4 (Web Text))

Max Activating Token Index: 485

Text #4

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.8268. Min Act: -0.1700

Data Index: 1238820 (C4 (Web Text))

Max Activating Token Index: 959

Text #5

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7955. Min Act: -0.1700

Data Index: 594525 (C4 (Web Text))

Max Activating Token Index: 393

Text #6

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7643. Min Act: -0.1700

Data Index: 937712 (C4 (Web Text))

Max Activating Token Index: 155

Text #7

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7643. Min Act: -0.1700

Data Index: 818689 (C4 (Web Text))

Max Activating Token Index: 679

Text #8

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7643. Min Act: -0.1700

Data Index: 706669 (C4 (Web Text))

Max Activating Token Index: 501

Text #9

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7643. Min Act: -0.1700

Data Index: 322964 (C4 (Web Text))

Max Activating Token Index: 903

Text #10

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7018. Min Act: -0.1700

Data Index: 773802 (C4 (Web Text))

Max Activating Token Index: 917

Text #11

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7330. Min Act: -0.1700

Data Index: 1320173 (C4 (Web Text))

Max Activating Token Index: 252

Text #12

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.6705. Min Act: -0.1700

Data Index: 865181 (C4 (Web Text))

Max Activating Token Index: 907

Text #13

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7018. Min Act: -0.1700

Data Index: 610098 (C4 (Web Text))

Max Activating Token Index: 626

Text #14

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7018. Min Act: -0.1700

Data Index: 741318 (C4 (Web Text))

Max Activating Token Index: 353

Text #15

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.6705. Min Act: -0.1700

Data Index: 1192944 (C4 (Web Text))

Max Activating Token Index: 370

Text #16

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.6705. Min Act: -0.1700

Data Index: 167575 (C4 (Web Text))

Max Activating Token Index: 462

Text #17

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.6705. Min Act: -0.1700

Data Index: 849174 (C4 (Web Text))

Max Activating Token Index: 735

Text #18

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.7018. Min Act: -0.1700

Data Index: 1340720 (C4 (Web Text))

Max Activating Token Index: 103

Text #19

Max Range: 5.9830. Min Range: -5.9830

Max Act: 5.6393. Min Act: -0.1700

Data Index: 366879 (C4 (Web Text))

Max Activating Token Index: 415