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

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

Neuron 728 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.0141. Min Range: -5.0141

Max Act: 5.0141. Min Act: -0.1700

Data Index: 1533131 (Python Code)

Max Activating Token Index: 242

Text #1

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.8578. Min Act: -0.1695

Data Index: 35434 (C4 (Web Text))

Max Activating Token Index: 218

Text #2

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.8891. Min Act: -0.1700

Data Index: 1181885 (C4 (Web Text))

Max Activating Token Index: 38

Text #3

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.8891. Min Act: -0.1700

Data Index: 459445 (C4 (Web Text))

Max Activating Token Index: 411

Text #4

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.8578. Min Act: -0.1700

Data Index: 664707 (C4 (Web Text))

Max Activating Token Index: 956

Text #5

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7953. Min Act: -0.1700

Data Index: 301508 (C4 (Web Text))

Max Activating Token Index: 509

Text #6

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7953. Min Act: -0.1697

Data Index: 1074826 (C4 (Web Text))

Max Activating Token Index: 744

Text #7

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7953. Min Act: -0.1700

Data Index: 721308 (C4 (Web Text))

Max Activating Token Index: 928

Text #8

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7641. Min Act: -0.1700

Data Index: 363947 (C4 (Web Text))

Max Activating Token Index: 33

Text #9

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7328. Min Act: -0.1700

Data Index: 358653 (C4 (Web Text))

Max Activating Token Index: 978

Text #10

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7016. Min Act: -0.1700

Data Index: 1181775 (C4 (Web Text))

Max Activating Token Index: 877

Text #11

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7016. Min Act: -0.1700

Data Index: 1204806 (C4 (Web Text))

Max Activating Token Index: 459

Text #12

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.7016. Min Act: -0.1700

Data Index: 906282 (C4 (Web Text))

Max Activating Token Index: 106

Text #13

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6703. Min Act: -0.1700

Data Index: 591512 (C4 (Web Text))

Max Activating Token Index: 586

Text #14

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6703. Min Act: -0.1700

Data Index: 1329142 (C4 (Web Text))

Max Activating Token Index: 288

Text #15

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6703. Min Act: -0.1700

Data Index: 439077 (C4 (Web Text))

Max Activating Token Index: 767

Text #16

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6703. Min Act: -0.1700

Data Index: 635694 (C4 (Web Text))

Max Activating Token Index: 82

Text #17

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6703. Min Act: -0.1700

Data Index: 189504 (C4 (Web Text))

Max Activating Token Index: 958

Text #18

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6703. Min Act: -0.1700

Data Index: 754091 (C4 (Web Text))

Max Activating Token Index: 477

Text #19

Max Range: 5.0141. Min Range: -5.0141

Max Act: 4.6391. Min Act: -0.1699

Data Index: 1174128 (C4 (Web Text))

Max Activating Token Index: 998