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

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

Neuron 1395 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.0762. Min Range: -3.0762

Max Act: 3.0762. Min Act: -0.1694

Data Index: 1147683 (C4 (Web Text))

Max Activating Token Index: 580

Text #1

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9972. Min Act: -0.1698

Data Index: 446604 (C4 (Web Text))

Max Activating Token Index: 813

Text #2

Max Range: 3.0762. Min Range: -3.0762

Max Act: 3.0131. Min Act: -0.1700

Data Index: 306843 (C4 (Web Text))

Max Activating Token Index: 904

Text #3

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9814. Min Act: -0.1700

Data Index: 889769 (C4 (Web Text))

Max Activating Token Index: 492

Text #4

Max Range: 3.0762. Min Range: -3.0762

Max Act: 3.0131. Min Act: -0.1699

Data Index: 92254 (C4 (Web Text))

Max Activating Token Index: 605

Text #5

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9814. Min Act: -0.1700

Data Index: 1407637 (Python Code)

Max Activating Token Index: 96

Text #6

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9656. Min Act: -0.1700

Data Index: 1639327 (Python Code)

Max Activating Token Index: 948

Text #7

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9972. Min Act: -0.1696

Data Index: 746120 (C4 (Web Text))

Max Activating Token Index: 359

Text #8

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9814. Min Act: -0.1699

Data Index: 239963 (C4 (Web Text))

Max Activating Token Index: 428

Text #9

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9656. Min Act: -0.1699

Data Index: 948315 (C4 (Web Text))

Max Activating Token Index: 780

Text #10

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9181. Min Act: -0.1700

Data Index: 1230297 (C4 (Web Text))

Max Activating Token Index: 614

Text #11

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9339. Min Act: -0.1696

Data Index: 174405 (C4 (Web Text))

Max Activating Token Index: 988

Text #12

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9498. Min Act: -0.1700

Data Index: 1399770 (Python Code)

Max Activating Token Index: 289

Text #13

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9339. Min Act: -0.1699

Data Index: 744936 (C4 (Web Text))

Max Activating Token Index: 652

Text #14

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9181. Min Act: -0.1700

Data Index: 1529791 (Python Code)

Max Activating Token Index: 1019

Text #15

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9339. Min Act: -0.1698

Data Index: 530701 (C4 (Web Text))

Max Activating Token Index: 902

Text #16

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.9022. Min Act: -0.1698

Data Index: 650867 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.8705. Min Act: -0.1700

Data Index: 1470037 (Python Code)

Max Activating Token Index: 1020

Text #18

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.8705. Min Act: -0.1699

Data Index: 581470 (C4 (Web Text))

Max Activating Token Index: 761

Text #19

Max Range: 3.0762. Min Range: -3.0762

Max Act: 2.8864. Min Act: -0.1700

Data Index: 1045795 (C4 (Web Text))

Max Activating Token Index: 323