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

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

Neuron 87 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: 3.4357. Min Range: -3.4357

Max Act: 3.4357. Min Act: -0.1700

Data Index: 1537772 (Python Code)

Max Activating Token Index: 451

Text #1

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.3886. Min Act: -0.1700

Data Index: 1377777 (Python Code)

Max Activating Token Index: 143

Text #2

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.3729. Min Act: -0.1700

Data Index: 1133853 (C4 (Web Text))

Max Activating Token Index: 387

Text #3

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2944. Min Act: -0.1700

Data Index: 165515 (C4 (Web Text))

Max Activating Token Index: 479

Text #4

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.3102. Min Act: -0.1700

Data Index: 835753 (C4 (Web Text))

Max Activating Token Index: 394

Text #5

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.3259. Min Act: -0.1700

Data Index: 1184196 (C4 (Web Text))

Max Activating Token Index: 355

Text #6

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.3102. Min Act: -0.1700

Data Index: 1575609 (Python Code)

Max Activating Token Index: 185

Text #7

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.3572. Min Act: -0.1700

Data Index: 469525 (C4 (Web Text))

Max Activating Token Index: 392

Text #8

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2944. Min Act: -0.1700

Data Index: 1181434 (C4 (Web Text))

Max Activating Token Index: 331

Text #9

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2630. Min Act: -0.1700

Data Index: 1643086 (Python Code)

Max Activating Token Index: 452

Text #10

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2787. Min Act: -0.1700

Data Index: 317104 (C4 (Web Text))

Max Activating Token Index: 377

Text #11

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2787. Min Act: -0.1700

Data Index: 560982 (C4 (Web Text))

Max Activating Token Index: 332

Text #12

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2787. Min Act: -0.1700

Data Index: 1101137 (C4 (Web Text))

Max Activating Token Index: 716

Text #13

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2630. Min Act: -0.1700

Data Index: 286191 (C4 (Web Text))

Max Activating Token Index: 501

Text #14

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2787. Min Act: -0.1700

Data Index: 1503859 (Python Code)

Max Activating Token Index: 139

Text #15

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2473. Min Act: -0.1700

Data Index: 1589792 (Python Code)

Max Activating Token Index: 42

Text #16

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2630. Min Act: -0.1700

Data Index: 1395668 (Python Code)

Max Activating Token Index: 152

Text #17

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2473. Min Act: -0.1700

Data Index: 654257 (C4 (Web Text))

Max Activating Token Index: 88

Text #18

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2630. Min Act: -0.1700

Data Index: 76021 (C4 (Web Text))

Max Activating Token Index: 660

Text #19

Max Range: 3.4357. Min Range: -3.4357

Max Act: 3.2316. Min Act: -0.1700

Data Index: 542350 (C4 (Web Text))

Max Activating Token Index: 266