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

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

Neuron 1940 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: 5.6067. Min Range: -5.6067

Max Act: 5.6067. Min Act: -0.1700

Data Index: 199588 (C4 (Web Text))

Max Activating Token Index: 985

Text #1

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.5442. Min Act: -0.1700

Data Index: 337484 (C4 (Web Text))

Max Activating Token Index: 515

Text #2

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.5442. Min Act: -0.1700

Data Index: 16986 (C4 (Web Text))

Max Activating Token Index: 568

Text #3

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.4817. Min Act: -0.1700

Data Index: 545983 (C4 (Web Text))

Max Activating Token Index: 567

Text #4

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.4505. Min Act: -0.1700

Data Index: 486548 (C4 (Web Text))

Max Activating Token Index: 566

Text #5

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.4505. Min Act: -0.1700

Data Index: 1220546 (C4 (Web Text))

Max Activating Token Index: 416

Text #6

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3880. Min Act: -0.1700

Data Index: 679622 (C4 (Web Text))

Max Activating Token Index: 735

Text #7

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3880. Min Act: -0.1700

Data Index: 1206520 (C4 (Web Text))

Max Activating Token Index: 391

Text #8

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3567. Min Act: -0.1700

Data Index: 79991 (C4 (Web Text))

Max Activating Token Index: 460

Text #9

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.2630. Min Act: -0.1700

Data Index: 912578 (C4 (Web Text))

Max Activating Token Index: 563

Text #10

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3255. Min Act: -0.1700

Data Index: 1114106 (C4 (Web Text))

Max Activating Token Index: 526

Text #11

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3255. Min Act: -0.1700

Data Index: 1100494 (C4 (Web Text))

Max Activating Token Index: 559

Text #12

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3567. Min Act: -0.1700

Data Index: 836118 (C4 (Web Text))

Max Activating Token Index: 939

Text #13

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3255. Min Act: -0.1700

Data Index: 561916 (C4 (Web Text))

Max Activating Token Index: 831

Text #14

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.3255. Min Act: -0.1700

Data Index: 786019 (C4 (Web Text))

Max Activating Token Index: 822

Text #15

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.2942. Min Act: -0.1700

Data Index: 195974 (C4 (Web Text))

Max Activating Token Index: 960

Text #16

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.2630. Min Act: -0.1700

Data Index: 1355708 (C4 (Web Text))

Max Activating Token Index: 434

Text #17

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.2317. Min Act: -0.1700

Data Index: 663030 (C4 (Web Text))

Max Activating Token Index: 592

Text #18

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.2005. Min Act: -0.1700

Data Index: 551193 (C4 (Web Text))

Max Activating Token Index: 772

Text #19

Max Range: 5.6067. Min Range: -5.6067

Max Act: 5.2630. Min Act: -0.1700

Data Index: 1157100 (C4 (Web Text))

Max Activating Token Index: 639