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

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

Neuron 1815 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.7032. Min Act: -0.1700

Data Index: 6049 (C4 (Web Text))

Max Activating Token Index: 645

Text #1

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.6719. Min Act: -0.1700

Data Index: 736778 (C4 (Web Text))

Max Activating Token Index: 545

Text #2

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.6249. Min Act: -0.1700

Data Index: 518099 (C4 (Web Text))

Max Activating Token Index: 986

Text #3

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.6093. Min Act: -0.1700

Data Index: 971896 (C4 (Web Text))

Max Activating Token Index: 724

Text #4

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.6249. Min Act: -0.1700

Data Index: 1101252 (C4 (Web Text))

Max Activating Token Index: 833

Text #5

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.6249. Min Act: -0.1700

Data Index: 613472 (C4 (Web Text))

Max Activating Token Index: 945

Text #6

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.5623. Min Act: -0.1700

Data Index: 518583 (C4 (Web Text))

Max Activating Token Index: 928

Text #7

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.5623. Min Act: -0.1700

Data Index: 1205725 (C4 (Web Text))

Max Activating Token Index: 36

Text #8

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.5779. Min Act: -0.1700

Data Index: 1109243 (C4 (Web Text))

Max Activating Token Index: 712

Text #9

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.5779. Min Act: -0.1700

Data Index: 1196345 (C4 (Web Text))

Max Activating Token Index: 934

Text #10

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.5153. Min Act: -0.1700

Data Index: 983660 (C4 (Web Text))

Max Activating Token Index: 998

Text #11

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4996. Min Act: -0.1700

Data Index: 1076846 (C4 (Web Text))

Max Activating Token Index: 31

Text #12

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4840. Min Act: -0.1700

Data Index: 569586 (C4 (Web Text))

Max Activating Token Index: 854

Text #13

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4526. Min Act: -0.1700

Data Index: 1399787 (Python Code)

Max Activating Token Index: 774

Text #14

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4996. Min Act: -0.1700

Data Index: 869535 (C4 (Web Text))

Max Activating Token Index: 37

Text #15

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4683. Min Act: -0.1700

Data Index: 538455 (C4 (Web Text))

Max Activating Token Index: 462

Text #16

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4683. Min Act: -0.1700

Data Index: 585378 (C4 (Web Text))

Max Activating Token Index: 818

Text #17

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4840. Min Act: -0.1700

Data Index: 757395 (C4 (Web Text))

Max Activating Token Index: 216

Text #18

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.4526. Min Act: -0.1700

Data Index: 927380 (C4 (Web Text))

Max Activating Token Index: 700

Text #19

Max Range: 3.7032. Min Range: -3.7032

Max Act: 3.5153. Min Act: -0.1700

Data Index: 898740 (C4 (Web Text))

Max Activating Token Index: 665