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

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

Neuron 670 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: 5.1367. Min Range: -5.1367

Max Act: 5.1367. Min Act: -0.1700

Data Index: 613741 (C4 (Web Text))

Max Activating Token Index: 746

Text #1

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.9805. Min Act: -0.1700

Data Index: 607983 (C4 (Web Text))

Max Activating Token Index: 237

Text #2

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.9180. Min Act: -0.1700

Data Index: 481892 (C4 (Web Text))

Max Activating Token Index: 700

Text #3

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.7617. Min Act: -0.1700

Data Index: 1347541 (C4 (Web Text))

Max Activating Token Index: 663

Text #4

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.7617. Min Act: -0.1700

Data Index: 1254267 (C4 (Web Text))

Max Activating Token Index: 968

Text #5

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.7617. Min Act: -0.1700

Data Index: 1035029 (C4 (Web Text))

Max Activating Token Index: 508

Text #6

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.7305. Min Act: -0.1700

Data Index: 42824 (C4 (Web Text))

Max Activating Token Index: 377

Text #7

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6680. Min Act: -0.1700

Data Index: 539542 (C4 (Web Text))

Max Activating Token Index: 868

Text #8

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.7305. Min Act: -0.1700

Data Index: 1246017 (C4 (Web Text))

Max Activating Token Index: 986

Text #9

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.7305. Min Act: -0.1700

Data Index: 226646 (C4 (Web Text))

Max Activating Token Index: 99

Text #10

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6992. Min Act: -0.1700

Data Index: 1297753 (C4 (Web Text))

Max Activating Token Index: 700

Text #11

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6367. Min Act: -0.1700

Data Index: 978996 (C4 (Web Text))

Max Activating Token Index: 634

Text #12

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6680. Min Act: -0.1700

Data Index: 105989 (C4 (Web Text))

Max Activating Token Index: 611

Text #13

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6992. Min Act: -0.1700

Data Index: 79727 (C4 (Web Text))

Max Activating Token Index: 652

Text #14

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6680. Min Act: -0.1700

Data Index: 1246579 (C4 (Web Text))

Max Activating Token Index: 941

Text #15

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6992. Min Act: -0.1700

Data Index: 230476 (C4 (Web Text))

Max Activating Token Index: 883

Text #16

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.5742. Min Act: -0.1700

Data Index: 300631 (C4 (Web Text))

Max Activating Token Index: 111

Text #17

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6367. Min Act: -0.1700

Data Index: 1208959 (C4 (Web Text))

Max Activating Token Index: 204

Text #18

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6680. Min Act: -0.1700

Data Index: 90938 (C4 (Web Text))

Max Activating Token Index: 825

Text #19

Max Range: 5.1367. Min Range: -5.1367

Max Act: 4.6367. Min Act: -0.1700

Data Index: 279193 (C4 (Web Text))

Max Activating Token Index: 929