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

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

Neuron 763 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: 3.4893. Min Range: -3.4893

Max Act: 3.4893. Min Act: -0.1700

Data Index: 1022072 (C4 (Web Text))

Max Activating Token Index: 994

Text #1

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4737. Min Act: -0.1700

Data Index: 1120972 (C4 (Web Text))

Max Activating Token Index: 828

Text #2

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4737. Min Act: -0.1700

Data Index: 1485669 (Python Code)

Max Activating Token Index: 110

Text #3

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4110. Min Act: -0.1700

Data Index: 387308 (C4 (Web Text))

Max Activating Token Index: 696

Text #4

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4580. Min Act: -0.1700

Data Index: 202893 (C4 (Web Text))

Max Activating Token Index: 496

Text #5

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4423. Min Act: -0.1700

Data Index: 646456 (C4 (Web Text))

Max Activating Token Index: 325

Text #6

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4580. Min Act: -0.1700

Data Index: 736848 (C4 (Web Text))

Max Activating Token Index: 71

Text #7

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4266. Min Act: -0.1700

Data Index: 961850 (C4 (Web Text))

Max Activating Token Index: 586

Text #8

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4266. Min Act: -0.1700

Data Index: 433761 (C4 (Web Text))

Max Activating Token Index: 797

Text #9

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4266. Min Act: -0.1700

Data Index: 315803 (C4 (Web Text))

Max Activating Token Index: 21

Text #10

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4266. Min Act: -0.1700

Data Index: 1238009 (C4 (Web Text))

Max Activating Token Index: 173

Text #11

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3953. Min Act: -0.1700

Data Index: 966492 (C4 (Web Text))

Max Activating Token Index: 593

Text #12

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.4266. Min Act: -0.1700

Data Index: 495747 (C4 (Web Text))

Max Activating Token Index: 21

Text #13

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3953. Min Act: -0.1700

Data Index: 934069 (C4 (Web Text))

Max Activating Token Index: 684

Text #14

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3796. Min Act: -0.1700

Data Index: 826798 (C4 (Web Text))

Max Activating Token Index: 202

Text #15

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3953. Min Act: -0.1700

Data Index: 1026260 (C4 (Web Text))

Max Activating Token Index: 656

Text #16

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3482. Min Act: -0.1700

Data Index: 1001197 (C4 (Web Text))

Max Activating Token Index: 707

Text #17

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3482. Min Act: -0.1700

Data Index: 823813 (C4 (Web Text))

Max Activating Token Index: 64

Text #18

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3639. Min Act: -0.1700

Data Index: 1024801 (C4 (Web Text))

Max Activating Token Index: 519

Text #19

Max Range: 3.4893. Min Range: -3.4893

Max Act: 3.3325. Min Act: -0.1700

Data Index: 913175 (C4 (Web Text))

Max Activating Token Index: 732