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

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

Neuron 127 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: 3.4538. Min Range: -3.4538

Max Act: 3.4538. Min Act: -0.1700

Data Index: 1359893 (Python Code)

Max Activating Token Index: 790

Text #1

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3754. Min Act: -0.1700

Data Index: 1014742 (C4 (Web Text))

Max Activating Token Index: 648

Text #2

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3597. Min Act: -0.1700

Data Index: 32751 (C4 (Web Text))

Max Activating Token Index: 545

Text #3

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3440. Min Act: -0.1700

Data Index: 61019 (C4 (Web Text))

Max Activating Token Index: 403

Text #4

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3754. Min Act: -0.1700

Data Index: 813945 (C4 (Web Text))

Max Activating Token Index: 412

Text #5

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3597. Min Act: -0.1700

Data Index: 1602007 (Python Code)

Max Activating Token Index: 621

Text #6

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3283. Min Act: -0.1700

Data Index: 1189370 (C4 (Web Text))

Max Activating Token Index: 146

Text #7

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3597. Min Act: -0.1700

Data Index: 529112 (C4 (Web Text))

Max Activating Token Index: 506

Text #8

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3126. Min Act: -0.1700

Data Index: 1333343 (C4 (Web Text))

Max Activating Token Index: 617

Text #9

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3440. Min Act: -0.1700

Data Index: 306573 (C4 (Web Text))

Max Activating Token Index: 432

Text #10

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3126. Min Act: -0.1700

Data Index: 154124 (C4 (Web Text))

Max Activating Token Index: 572

Text #11

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2969. Min Act: -0.1700

Data Index: 1230878 (C4 (Web Text))

Max Activating Token Index: 592

Text #12

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2812. Min Act: -0.1700

Data Index: 66679 (C4 (Web Text))

Max Activating Token Index: 530

Text #13

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.3597. Min Act: -0.1700

Data Index: 1074478 (C4 (Web Text))

Max Activating Token Index: 657

Text #14

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2969. Min Act: -0.1700

Data Index: 450274 (C4 (Web Text))

Max Activating Token Index: 661

Text #15

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2812. Min Act: -0.1700

Data Index: 1355666 (C4 (Web Text))

Max Activating Token Index: 578

Text #16

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2812. Min Act: -0.1700

Data Index: 346723 (C4 (Web Text))

Max Activating Token Index: 419

Text #17

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2655. Min Act: -0.1700

Data Index: 1344968 (C4 (Web Text))

Max Activating Token Index: 323

Text #18

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.2498. Min Act: -0.1700

Data Index: 1499235 (Python Code)

Max Activating Token Index: 638

Text #19

Max Range: 3.4538. Min Range: -3.4538

Max Act: 3.1868. Min Act: -0.1700

Data Index: 1583697 (Python Code)

Max Activating Token Index: 749