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

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

Neuron 1769 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: 3.1821. Min Range: -3.1821

Max Act: 3.1821. Min Act: -0.1700

Data Index: 1586713 (Python Code)

Max Activating Token Index: 795

Text #1

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.1348. Min Act: -0.1700

Data Index: 1321981 (C4 (Web Text))

Max Activating Token Index: 819

Text #2

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.1033. Min Act: -0.1700

Data Index: 1572407 (Python Code)

Max Activating Token Index: 480

Text #3

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.1033. Min Act: -0.1700

Data Index: 1603292 (Python Code)

Max Activating Token Index: 462

Text #4

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0718. Min Act: -0.1694

Data Index: 1612139 (Python Code)

Max Activating Token Index: 200

Text #5

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0402. Min Act: -0.1700

Data Index: 1620993 (Python Code)

Max Activating Token Index: 707

Text #6

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0718. Min Act: -0.1700

Data Index: 1634058 (Python Code)

Max Activating Token Index: 434

Text #7

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0402. Min Act: -0.1700

Data Index: 1420283 (Python Code)

Max Activating Token Index: 798

Text #8

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0244. Min Act: -0.1700

Data Index: 1429345 (Python Code)

Max Activating Token Index: 216

Text #9

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0086. Min Act: -0.1698

Data Index: 1474623 (Python Code)

Max Activating Token Index: 600

Text #10

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0244. Min Act: -0.1700

Data Index: 1502693 (Python Code)

Max Activating Token Index: 468

Text #11

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9928. Min Act: -0.1684

Data Index: 1646691 (Python Code)

Max Activating Token Index: 300

Text #12

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9453. Min Act: -0.1700

Data Index: 1526132 (Python Code)

Max Activating Token Index: 803

Text #13

Max Range: 3.1821. Min Range: -3.1821

Max Act: 3.0086. Min Act: -0.1700

Data Index: 1362105 (Python Code)

Max Activating Token Index: 945

Text #14

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9928. Min Act: -0.1697

Data Index: 1438422 (Python Code)

Max Activating Token Index: 163

Text #15

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9928. Min Act: -0.1700

Data Index: 1414267 (Python Code)

Max Activating Token Index: 594

Text #16

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9770. Min Act: -0.1699

Data Index: 1598127 (Python Code)

Max Activating Token Index: 599

Text #17

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9770. Min Act: -0.1700

Data Index: 1528064 (Python Code)

Max Activating Token Index: 740

Text #18

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9770. Min Act: -0.1700

Data Index: 602920 (C4 (Web Text))

Max Activating Token Index: 119

Text #19

Max Range: 3.1821. Min Range: -3.1821

Max Act: 2.9770. Min Act: -0.1696

Data Index: 1541313 (Python Code)

Max Activating Token Index: 596