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

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

Neuron 1205 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.7759. Min Act: -0.1700

Data Index: 1534012 (Python Code)

Max Activating Token Index: 761

Text #1

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.6822. Min Act: -0.1700

Data Index: 606738 (C4 (Web Text))

Max Activating Token Index: 753

Text #2

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.6822. Min Act: -0.1700

Data Index: 647548 (C4 (Web Text))

Max Activating Token Index: 142

Text #3

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.6197. Min Act: -0.1700

Data Index: 1636864 (Python Code)

Max Activating Token Index: 361

Text #4

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.5572. Min Act: -0.1700

Data Index: 101364 (C4 (Web Text))

Max Activating Token Index: 781

Text #5

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.5572. Min Act: -0.1700

Data Index: 920473 (C4 (Web Text))

Max Activating Token Index: 244

Text #6

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4947. Min Act: -0.1700

Data Index: 305605 (C4 (Web Text))

Max Activating Token Index: 351

Text #7

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.5259. Min Act: -0.1700

Data Index: 1338386 (C4 (Web Text))

Max Activating Token Index: 704

Text #8

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4947. Min Act: -0.1700

Data Index: 1315223 (C4 (Web Text))

Max Activating Token Index: 554

Text #9

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4947. Min Act: -0.1700

Data Index: 708272 (C4 (Web Text))

Max Activating Token Index: 660

Text #10

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.5259. Min Act: -0.1700

Data Index: 1518810 (Python Code)

Max Activating Token Index: 992

Text #11

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.5259. Min Act: -0.1700

Data Index: 678959 (C4 (Web Text))

Max Activating Token Index: 419

Text #12

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4947. Min Act: -0.1700

Data Index: 1519652 (Python Code)

Max Activating Token Index: 692

Text #13

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4634. Min Act: -0.1700

Data Index: 1091891 (C4 (Web Text))

Max Activating Token Index: 765

Text #14

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4947. Min Act: -0.1700

Data Index: 86426 (C4 (Web Text))

Max Activating Token Index: 268

Text #15

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4634. Min Act: -0.1700

Data Index: 1078202 (C4 (Web Text))

Max Activating Token Index: 831

Text #16

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4322. Min Act: -0.1700

Data Index: 362908 (C4 (Web Text))

Max Activating Token Index: 905

Text #17

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4322. Min Act: -0.1700

Data Index: 1395710 (Python Code)

Max Activating Token Index: 945

Text #18

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4322. Min Act: -0.1700

Data Index: 1530564 (Python Code)

Max Activating Token Index: 586

Text #19

Max Range: 5.7759. Min Range: -5.7759

Max Act: 5.4322. Min Act: -0.1700

Data Index: 1473753 (Python Code)

Max Activating Token Index: 908