Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1048 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.9903. Min Act: -0.1700

Data Index: 1284565 (C4 (Web Text))

Max Activating Token Index: 959

Text #1

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.9278. Min Act: -0.1700

Data Index: 1452772 (Python Code)

Max Activating Token Index: 500

Text #2

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.8340. Min Act: -0.1700

Data Index: 1065371 (C4 (Web Text))

Max Activating Token Index: 914

Text #3

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.7403. Min Act: -0.1700

Data Index: 537980 (C4 (Web Text))

Max Activating Token Index: 964

Text #4

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.8028. Min Act: -0.1700

Data Index: 387199 (C4 (Web Text))

Max Activating Token Index: 726

Text #5

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.8028. Min Act: -0.1700

Data Index: 1213712 (C4 (Web Text))

Max Activating Token Index: 579

Text #6

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.7403. Min Act: -0.1700

Data Index: 942374 (C4 (Web Text))

Max Activating Token Index: 705

Text #7

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.7403. Min Act: -0.1700

Data Index: 1396218 (Python Code)

Max Activating Token Index: 248

Text #8

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6778. Min Act: -0.1700

Data Index: 415820 (C4 (Web Text))

Max Activating Token Index: 931

Text #9

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6465. Min Act: -0.1700

Data Index: 1289684 (C4 (Web Text))

Max Activating Token Index: 692

Text #10

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.7090. Min Act: -0.1700

Data Index: 442440 (C4 (Web Text))

Max Activating Token Index: 836

Text #11

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.5528. Min Act: -0.1700

Data Index: 1025859 (C4 (Web Text))

Max Activating Token Index: 834

Text #12

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6778. Min Act: -0.1700

Data Index: 551958 (C4 (Web Text))

Max Activating Token Index: 950

Text #13

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6153. Min Act: -0.1700

Data Index: 729379 (C4 (Web Text))

Max Activating Token Index: 611

Text #14

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6153. Min Act: -0.1700

Data Index: 618427 (C4 (Web Text))

Max Activating Token Index: 518

Text #15

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6153. Min Act: -0.1700

Data Index: 1125601 (C4 (Web Text))

Max Activating Token Index: 268

Text #16

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6465. Min Act: -0.1700

Data Index: 200296 (C4 (Web Text))

Max Activating Token Index: 201

Text #17

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.6153. Min Act: -0.1700

Data Index: 163921 (C4 (Web Text))

Max Activating Token Index: 546

Text #18

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.5840. Min Act: -0.1700

Data Index: 190957 (C4 (Web Text))

Max Activating Token Index: 937

Text #19

Max Range: 5.9903. Min Range: -5.9903

Max Act: 5.5840. Min Act: -0.1700

Data Index: 974635 (C4 (Web Text))

Max Activating Token Index: 802