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

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

Neuron 1048 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: 3.7418. Min Range: -3.7418

Max Act: 3.7418. Min Act: -0.1700

Data Index: 967211 (C4 (Web Text))

Max Activating Token Index: 845

Text #1

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.6636. Min Act: -0.1700

Data Index: 824910 (C4 (Web Text))

Max Activating Token Index: 230

Text #2

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.6009. Min Act: -0.1700

Data Index: 1236227 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5696. Min Act: -0.1700

Data Index: 310170 (C4 (Web Text))

Max Activating Token Index: 148

Text #4

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5853. Min Act: -0.1700

Data Index: 680957 (C4 (Web Text))

Max Activating Token Index: 801

Text #5

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5853. Min Act: -0.1700

Data Index: 401217 (C4 (Web Text))

Max Activating Token Index: 60

Text #6

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5383. Min Act: -0.1700

Data Index: 453827 (C4 (Web Text))

Max Activating Token Index: 828

Text #7

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5226. Min Act: -0.1700

Data Index: 1295252 (C4 (Web Text))

Max Activating Token Index: 958

Text #8

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5226. Min Act: -0.1700

Data Index: 612169 (C4 (Web Text))

Max Activating Token Index: 261

Text #9

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5070. Min Act: -0.1700

Data Index: 550104 (C4 (Web Text))

Max Activating Token Index: 777

Text #10

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5383. Min Act: -0.1700

Data Index: 883170 (C4 (Web Text))

Max Activating Token Index: 536

Text #11

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5070. Min Act: -0.1700

Data Index: 277272 (C4 (Web Text))

Max Activating Token Index: 801

Text #12

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.5070. Min Act: -0.1700

Data Index: 1377754 (Python Code)

Max Activating Token Index: 336

Text #13

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4756. Min Act: -0.1700

Data Index: 1306779 (C4 (Web Text))

Max Activating Token Index: 569

Text #14

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4756. Min Act: -0.1700

Data Index: 127816 (C4 (Web Text))

Max Activating Token Index: 895

Text #15

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4756. Min Act: -0.1700

Data Index: 931303 (C4 (Web Text))

Max Activating Token Index: 392

Text #16

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4756. Min Act: -0.1700

Data Index: 255209 (C4 (Web Text))

Max Activating Token Index: 401

Text #17

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4913. Min Act: -0.1700

Data Index: 297357 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4756. Min Act: -0.1700

Data Index: 735948 (C4 (Web Text))

Max Activating Token Index: 245

Text #19

Max Range: 3.7418. Min Range: -3.7418

Max Act: 3.4600. Min Act: -0.1700

Data Index: 198245 (C4 (Web Text))

Max Activating Token Index: 706