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

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

Neuron 1823 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: 3.9429. Min Range: -3.9429

Max Act: 3.9429. Min Act: -0.1700

Data Index: 338196 (C4 (Web Text))

Max Activating Token Index: 765

Text #1

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9429. Min Act: -0.1700

Data Index: 975663 (C4 (Web Text))

Max Activating Token Index: 516

Text #2

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 1435708 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 960216 (C4 (Web Text))

Max Activating Token Index: 752

Text #4

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 1604155 (Python Code)

Max Activating Token Index: 343

Text #5

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 1304286 (C4 (Web Text))

Max Activating Token Index: 334

Text #6

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 57233 (C4 (Web Text))

Max Activating Token Index: 518

Text #7

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 176237 (C4 (Web Text))

Max Activating Token Index: 304

Text #8

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 973432 (C4 (Web Text))

Max Activating Token Index: 515

Text #9

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8804. Min Act: -0.1700

Data Index: 1117575 (C4 (Web Text))

Max Activating Token Index: 514

Text #10

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8804. Min Act: -0.1700

Data Index: 1589571 (Python Code)

Max Activating Token Index: 575

Text #11

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8491. Min Act: -0.1700

Data Index: 1614106 (Python Code)

Max Activating Token Index: 507

Text #12

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8491. Min Act: -0.1700

Data Index: 1387765 (Python Code)

Max Activating Token Index: 549

Text #13

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8491. Min Act: -0.1700

Data Index: 639425 (C4 (Web Text))

Max Activating Token Index: 844

Text #14

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8804. Min Act: -0.1700

Data Index: 901042 (C4 (Web Text))

Max Activating Token Index: 217

Text #15

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8491. Min Act: -0.1700

Data Index: 386918 (C4 (Web Text))

Max Activating Token Index: 533

Text #16

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 1066131 (C4 (Web Text))

Max Activating Token Index: 751

Text #17

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.9116. Min Act: -0.1700

Data Index: 1150168 (C4 (Web Text))

Max Activating Token Index: 440

Text #18

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8804. Min Act: -0.1700

Data Index: 1206865 (C4 (Web Text))

Max Activating Token Index: 560

Text #19

Max Range: 3.9429. Min Range: -3.9429

Max Act: 3.8804. Min Act: -0.1700

Data Index: 1212007 (C4 (Web Text))

Max Activating Token Index: 412