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

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

Neuron 769 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: 9.0308. Min Range: -9.0308

Max Act: 9.0308. Min Act: -0.1699

Data Index: 1260440 (C4 (Web Text))

Max Activating Token Index: 169

Text #1

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.9058. Min Act: -0.1695

Data Index: 1300756 (C4 (Web Text))

Max Activating Token Index: 570

Text #2

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.7183. Min Act: -0.1700

Data Index: 429038 (C4 (Web Text))

Max Activating Token Index: 501

Text #3

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.8433. Min Act: -0.1699

Data Index: 554249 (C4 (Web Text))

Max Activating Token Index: 863

Text #4

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.5933. Min Act: -0.1697

Data Index: 1105112 (C4 (Web Text))

Max Activating Token Index: 461

Text #5

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.7183. Min Act: -0.1665

Data Index: 660140 (C4 (Web Text))

Max Activating Token Index: 334

Text #6

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.6558. Min Act: -0.1696

Data Index: 317353 (C4 (Web Text))

Max Activating Token Index: 20

Text #7

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.5933. Min Act: -0.1699

Data Index: 1018216 (C4 (Web Text))

Max Activating Token Index: 681

Text #8

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.7183. Min Act: -0.1673

Data Index: 1016115 (C4 (Web Text))

Max Activating Token Index: 729

Text #9

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.5308. Min Act: -0.1700

Data Index: 810380 (C4 (Web Text))

Max Activating Token Index: 885

Text #10

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4683. Min Act: -0.1638

Data Index: 693536 (C4 (Web Text))

Max Activating Token Index: 1006

Text #11

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4058. Min Act: -0.1700

Data Index: 1321621 (C4 (Web Text))

Max Activating Token Index: 581

Text #12

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4683. Min Act: -0.1697

Data Index: 918973 (C4 (Web Text))

Max Activating Token Index: 357

Text #13

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4058. Min Act: -0.1700

Data Index: 43421 (C4 (Web Text))

Max Activating Token Index: 71

Text #14

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4683. Min Act: -0.1696

Data Index: 18124 (C4 (Web Text))

Max Activating Token Index: 639

Text #15

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.1558. Min Act: -0.1700

Data Index: 343786 (C4 (Web Text))

Max Activating Token Index: 846

Text #16

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4683. Min Act: -0.1690

Data Index: 1037940 (C4 (Web Text))

Max Activating Token Index: 650

Text #17

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.2183. Min Act: -0.1700

Data Index: 1119139 (C4 (Web Text))

Max Activating Token Index: 946

Text #18

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.3433. Min Act: -0.1628

Data Index: 614172 (C4 (Web Text))

Max Activating Token Index: 25

Text #19

Max Range: 9.0308. Min Range: -9.0308

Max Act: 8.4683. Min Act: -0.1657

Data Index: 438232 (C4 (Web Text))

Max Activating Token Index: 203