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

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

Neuron 423 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: 3.4694. Min Range: -3.4694

Max Act: 3.4694. Min Act: -0.1700

Data Index: 1116441 (C4 (Web Text))

Max Activating Token Index: 424

Text #1

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.4067. Min Act: -0.1700

Data Index: 775730 (C4 (Web Text))

Max Activating Token Index: 237

Text #2

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.3910. Min Act: -0.1700

Data Index: 1292507 (C4 (Web Text))

Max Activating Token Index: 921

Text #3

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.3596. Min Act: -0.1700

Data Index: 709903 (C4 (Web Text))

Max Activating Token Index: 553

Text #4

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.3439. Min Act: -0.1700

Data Index: 417020 (C4 (Web Text))

Max Activating Token Index: 554

Text #5

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2811. Min Act: -0.1700

Data Index: 571654 (C4 (Web Text))

Max Activating Token Index: 872

Text #6

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2968. Min Act: -0.1700

Data Index: 966659 (C4 (Web Text))

Max Activating Token Index: 208

Text #7

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2654. Min Act: -0.1700

Data Index: 254550 (C4 (Web Text))

Max Activating Token Index: 486

Text #8

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2497. Min Act: -0.1700

Data Index: 1134549 (C4 (Web Text))

Max Activating Token Index: 804

Text #9

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2182. Min Act: -0.1700

Data Index: 1052820 (C4 (Web Text))

Max Activating Token Index: 269

Text #10

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2025. Min Act: -0.1700

Data Index: 159001 (C4 (Web Text))

Max Activating Token Index: 950

Text #11

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2182. Min Act: -0.1700

Data Index: 428102 (C4 (Web Text))

Max Activating Token Index: 757

Text #12

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2339. Min Act: -0.1700

Data Index: 943743 (C4 (Web Text))

Max Activating Token Index: 534

Text #13

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2339. Min Act: -0.1700

Data Index: 1067024 (C4 (Web Text))

Max Activating Token Index: 620

Text #14

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.1710. Min Act: -0.1700

Data Index: 602298 (C4 (Web Text))

Max Activating Token Index: 523

Text #15

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.1868. Min Act: -0.1700

Data Index: 149813 (C4 (Web Text))

Max Activating Token Index: 613

Text #16

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.1868. Min Act: -0.1700

Data Index: 179400 (C4 (Web Text))

Max Activating Token Index: 909

Text #17

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2025. Min Act: -0.1700

Data Index: 871751 (C4 (Web Text))

Max Activating Token Index: 543

Text #18

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2025. Min Act: -0.1700

Data Index: 1103272 (C4 (Web Text))

Max Activating Token Index: 465

Text #19

Max Range: 3.4694. Min Range: -3.4694

Max Act: 3.2182. Min Act: -0.1700

Data Index: 1651620 (Python Code)

Max Activating Token Index: 193