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

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

Neuron 1836 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: 5.3288. Min Range: -5.3288

Max Act: 5.3288. Min Act: -0.1700

Data Index: 1265785 (C4 (Web Text))

Max Activating Token Index: 43

Text #1

Max Range: 5.3288. Min Range: -5.3288

Max Act: 5.0476. Min Act: -0.1700

Data Index: 1137732 (C4 (Web Text))

Max Activating Token Index: 68

Text #2

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.9538. Min Act: -0.1700

Data Index: 687450 (C4 (Web Text))

Max Activating Token Index: 402

Text #3

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.9226. Min Act: -0.1700

Data Index: 507539 (C4 (Web Text))

Max Activating Token Index: 326

Text #4

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.9226. Min Act: -0.1700

Data Index: 389419 (C4 (Web Text))

Max Activating Token Index: 288

Text #5

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8601. Min Act: -0.1700

Data Index: 358310 (C4 (Web Text))

Max Activating Token Index: 518

Text #6

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 689805 (C4 (Web Text))

Max Activating Token Index: 819

Text #7

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 625166 (C4 (Web Text))

Max Activating Token Index: 352

Text #8

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 760386 (C4 (Web Text))

Max Activating Token Index: 780

Text #9

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 97129 (C4 (Web Text))

Max Activating Token Index: 330

Text #10

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 927665 (C4 (Web Text))

Max Activating Token Index: 34

Text #11

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.9226. Min Act: -0.1700

Data Index: 528391 (C4 (Web Text))

Max Activating Token Index: 901

Text #12

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 932740 (C4 (Web Text))

Max Activating Token Index: 387

Text #13

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8288. Min Act: -0.1700

Data Index: 706730 (C4 (Web Text))

Max Activating Token Index: 42

Text #14

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.7976. Min Act: -0.1700

Data Index: 1245231 (C4 (Web Text))

Max Activating Token Index: 403

Text #15

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8601. Min Act: -0.1700

Data Index: 97800 (C4 (Web Text))

Max Activating Token Index: 194

Text #16

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8288. Min Act: -0.1700

Data Index: 502529 (C4 (Web Text))

Max Activating Token Index: 729

Text #17

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.7976. Min Act: -0.1700

Data Index: 1321833 (C4 (Web Text))

Max Activating Token Index: 455

Text #18

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8601. Min Act: -0.1700

Data Index: 196520 (C4 (Web Text))

Max Activating Token Index: 120

Text #19

Max Range: 5.3288. Min Range: -5.3288

Max Act: 4.8913. Min Act: -0.1700

Data Index: 48349 (C4 (Web Text))

Max Activating Token Index: 34