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

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

Neuron 2024 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.2335. Min Act: -0.1700

Data Index: 1457079 (Python Code)

Max Activating Token Index: 687

Text #1

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.1398. Min Act: -0.1700

Data Index: 1496198 (Python Code)

Max Activating Token Index: 764

Text #2

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.1241. Min Act: -0.1700

Data Index: 1582726 (Python Code)

Max Activating Token Index: 289

Text #3

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.1241. Min Act: -0.1700

Data Index: 1571806 (Python Code)

Max Activating Token Index: 270

Text #4

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.9991. Min Act: -0.1700

Data Index: 1599378 (Python Code)

Max Activating Token Index: 328

Text #5

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.0929. Min Act: -0.1700

Data Index: 1620296 (Python Code)

Max Activating Token Index: 452

Text #6

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.1085. Min Act: -0.1700

Data Index: 1558480 (Python Code)

Max Activating Token Index: 297

Text #7

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.1398. Min Act: -0.1700

Data Index: 1655225 (Python Code)

Max Activating Token Index: 895

Text #8

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.8741. Min Act: -0.1700

Data Index: 1547279 (Python Code)

Max Activating Token Index: 487

Text #9

Max Range: 7.2335. Min Range: -7.2335

Max Act: 7.0148. Min Act: -0.1700

Data Index: 1603467 (Python Code)

Max Activating Token Index: 889

Text #10

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.9679. Min Act: -0.1700

Data Index: 1387573 (Python Code)

Max Activating Token Index: 331

Text #11

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.9366. Min Act: -0.1700

Data Index: 606129 (C4 (Web Text))

Max Activating Token Index: 834

Text #12

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.9210. Min Act: -0.1700

Data Index: 596787 (C4 (Web Text))

Max Activating Token Index: 532

Text #13

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.8273. Min Act: -0.1700

Data Index: 764504 (C4 (Web Text))

Max Activating Token Index: 496

Text #14

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.9210. Min Act: -0.1700

Data Index: 1608494 (Python Code)

Max Activating Token Index: 564

Text #15

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.8116. Min Act: -0.1700

Data Index: 1491295 (Python Code)

Max Activating Token Index: 192

Text #16

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.8898. Min Act: -0.1700

Data Index: 1553455 (Python Code)

Max Activating Token Index: 315

Text #17

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.9054. Min Act: -0.1700

Data Index: 1580388 (Python Code)

Max Activating Token Index: 268

Text #18

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.7726. Min Act: -0.1700

Data Index: 1620967 (Python Code)

Max Activating Token Index: 344

Text #19

Max Range: 7.2335. Min Range: -7.2335

Max Act: 6.8429. Min Act: -0.1700

Data Index: 1416923 (Python Code)

Max Activating Token Index: 424