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

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

Neuron 1571 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.6855. Min Act: -0.1700

Data Index: 779868 (C4 (Web Text))

Max Activating Token Index: 557

Text #1

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.6699. Min Act: -0.1700

Data Index: 906595 (C4 (Web Text))

Max Activating Token Index: 677

Text #2

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5759. Min Act: -0.1700

Data Index: 99395 (C4 (Web Text))

Max Activating Token Index: 567

Text #3

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.6073. Min Act: -0.1700

Data Index: 833453 (C4 (Web Text))

Max Activating Token Index: 947

Text #4

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.6073. Min Act: -0.1700

Data Index: 330140 (C4 (Web Text))

Max Activating Token Index: 271

Text #5

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5759. Min Act: -0.1700

Data Index: 204419 (C4 (Web Text))

Max Activating Token Index: 286

Text #6

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5603. Min Act: -0.1700

Data Index: 388347 (C4 (Web Text))

Max Activating Token Index: 643

Text #7

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5603. Min Act: -0.1700

Data Index: 640015 (C4 (Web Text))

Max Activating Token Index: 194

Text #8

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5603. Min Act: -0.1700

Data Index: 976637 (C4 (Web Text))

Max Activating Token Index: 311

Text #9

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5290. Min Act: -0.1700

Data Index: 552918 (C4 (Web Text))

Max Activating Token Index: 973

Text #10

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4976. Min Act: -0.1700

Data Index: 830667 (C4 (Web Text))

Max Activating Token Index: 380

Text #11

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4976. Min Act: -0.1700

Data Index: 540150 (C4 (Web Text))

Max Activating Token Index: 992

Text #12

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4820. Min Act: -0.1700

Data Index: 1355024 (C4 (Web Text))

Max Activating Token Index: 379

Text #13

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4663. Min Act: -0.1700

Data Index: 883777 (C4 (Web Text))

Max Activating Token Index: 745

Text #14

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.5133. Min Act: -0.1700

Data Index: 24862 (C4 (Web Text))

Max Activating Token Index: 463

Text #15

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4506. Min Act: -0.1700

Data Index: 402733 (C4 (Web Text))

Max Activating Token Index: 437

Text #16

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4663. Min Act: -0.1700

Data Index: 233557 (C4 (Web Text))

Max Activating Token Index: 453

Text #17

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4349. Min Act: -0.1700

Data Index: 849708 (C4 (Web Text))

Max Activating Token Index: 666

Text #18

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4506. Min Act: -0.1700

Data Index: 1096971 (C4 (Web Text))

Max Activating Token Index: 532

Text #19

Max Range: 3.6855. Min Range: -3.6855

Max Act: 3.4506. Min Act: -0.1700

Data Index: 481047 (C4 (Web Text))

Max Activating Token Index: 693