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

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

Neuron 825 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: 7.1196. Min Range: -7.1196

Max Act: 7.1196. Min Act: -0.1700

Data Index: 68388 (C4 (Web Text))

Max Activating Token Index: 777

Text #1

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.8071. Min Act: -0.1700

Data Index: 828685 (C4 (Web Text))

Max Activating Token Index: 1012

Text #2

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.9321. Min Act: -0.1700

Data Index: 951889 (C4 (Web Text))

Max Activating Token Index: 443

Text #3

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.7446. Min Act: -0.1700

Data Index: 1095233 (C4 (Web Text))

Max Activating Token Index: 851

Text #4

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.7133. Min Act: -0.1700

Data Index: 214909 (C4 (Web Text))

Max Activating Token Index: 390

Text #5

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.8071. Min Act: -0.1700

Data Index: 1127683 (C4 (Web Text))

Max Activating Token Index: 363

Text #6

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6508. Min Act: -0.1700

Data Index: 476147 (C4 (Web Text))

Max Activating Token Index: 385

Text #7

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6821. Min Act: -0.1700

Data Index: 144786 (C4 (Web Text))

Max Activating Token Index: 922

Text #8

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.7758. Min Act: -0.1700

Data Index: 959425 (C4 (Web Text))

Max Activating Token Index: 338

Text #9

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.7446. Min Act: -0.1700

Data Index: 805674 (C4 (Web Text))

Max Activating Token Index: 978

Text #10

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.7758. Min Act: -0.1700

Data Index: 892757 (C4 (Web Text))

Max Activating Token Index: 924

Text #11

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.5883. Min Act: -0.1700

Data Index: 1341147 (C4 (Web Text))

Max Activating Token Index: 964

Text #12

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6821. Min Act: -0.1700

Data Index: 517070 (C4 (Web Text))

Max Activating Token Index: 498

Text #13

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.5883. Min Act: -0.1700

Data Index: 831954 (C4 (Web Text))

Max Activating Token Index: 825

Text #14

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6821. Min Act: -0.1700

Data Index: 1082121 (C4 (Web Text))

Max Activating Token Index: 496

Text #15

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.4946. Min Act: -0.1700

Data Index: 696807 (C4 (Web Text))

Max Activating Token Index: 414

Text #16

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.7758. Min Act: -0.1700

Data Index: 528026 (C4 (Web Text))

Max Activating Token Index: 455

Text #17

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6196. Min Act: -0.1700

Data Index: 97608 (C4 (Web Text))

Max Activating Token Index: 979

Text #18

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6196. Min Act: -0.1700

Data Index: 481323 (C4 (Web Text))

Max Activating Token Index: 668

Text #19

Max Range: 7.1196. Min Range: -7.1196

Max Act: 6.6196. Min Act: -0.1700

Data Index: 331286 (C4 (Web Text))

Max Activating Token Index: 87