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

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

Neuron 529 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.4529. Min Act: -0.1700

Data Index: 996071 (C4 (Web Text))

Max Activating Token Index: 332

Text #1

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.3902. Min Act: -0.1700

Data Index: 775665 (C4 (Web Text))

Max Activating Token Index: 712

Text #2

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.3902. Min Act: -0.1700

Data Index: 668942 (C4 (Web Text))

Max Activating Token Index: 766

Text #3

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.3431. Min Act: -0.1700

Data Index: 910729 (C4 (Web Text))

Max Activating Token Index: 79

Text #4

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.3117. Min Act: -0.1700

Data Index: 289473 (C4 (Web Text))

Max Activating Token Index: 846

Text #5

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2960. Min Act: -0.1700

Data Index: 687039 (C4 (Web Text))

Max Activating Token Index: 997

Text #6

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2803. Min Act: -0.1700

Data Index: 734131 (C4 (Web Text))

Max Activating Token Index: 120

Text #7

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2803. Min Act: -0.1700

Data Index: 1308683 (C4 (Web Text))

Max Activating Token Index: 711

Text #8

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2489. Min Act: -0.1700

Data Index: 1308525 (C4 (Web Text))

Max Activating Token Index: 242

Text #9

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2332. Min Act: -0.1700

Data Index: 170619 (C4 (Web Text))

Max Activating Token Index: 747

Text #10

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2332. Min Act: -0.1700

Data Index: 1194643 (C4 (Web Text))

Max Activating Token Index: 978

Text #11

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2174. Min Act: -0.1700

Data Index: 346818 (C4 (Web Text))

Max Activating Token Index: 810

Text #12

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2017. Min Act: -0.1700

Data Index: 593287 (C4 (Web Text))

Max Activating Token Index: 278

Text #13

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2017. Min Act: -0.1700

Data Index: 1597196 (Python Code)

Max Activating Token Index: 825

Text #14

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.2017. Min Act: -0.1700

Data Index: 547236 (C4 (Web Text))

Max Activating Token Index: 751

Text #15

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.1860. Min Act: -0.1700

Data Index: 334024 (C4 (Web Text))

Max Activating Token Index: 881

Text #16

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.1545. Min Act: -0.1700

Data Index: 1295264 (C4 (Web Text))

Max Activating Token Index: 20

Text #17

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.1702. Min Act: -0.1700

Data Index: 1163570 (C4 (Web Text))

Max Activating Token Index: 844

Text #18

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.1702. Min Act: -0.1700

Data Index: 1237446 (C4 (Web Text))

Max Activating Token Index: 969

Text #19

Max Range: 3.4529. Min Range: -3.4529

Max Act: 3.1860. Min Act: -0.1700

Data Index: 482690 (C4 (Web Text))

Max Activating Token Index: 96