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

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

Neuron 1346 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.8567. Min Act: -0.1700

Data Index: 845501 (C4 (Web Text))

Max Activating Token Index: 952

Text #1

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.8880. Min Act: -0.1700

Data Index: 704673 (C4 (Web Text))

Max Activating Token Index: 907

Text #2

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.8255. Min Act: -0.1700

Data Index: 835700 (C4 (Web Text))

Max Activating Token Index: 694

Text #3

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7942. Min Act: -0.1700

Data Index: 804382 (C4 (Web Text))

Max Activating Token Index: 479

Text #4

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.8255. Min Act: -0.1700

Data Index: 846539 (C4 (Web Text))

Max Activating Token Index: 513

Text #5

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7629. Min Act: -0.1700

Data Index: 478817 (C4 (Web Text))

Max Activating Token Index: 888

Text #6

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7629. Min Act: -0.1700

Data Index: 1273756 (C4 (Web Text))

Max Activating Token Index: 952

Text #7

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7629. Min Act: -0.1700

Data Index: 280522 (C4 (Web Text))

Max Activating Token Index: 765

Text #8

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 1302125 (C4 (Web Text))

Max Activating Token Index: 193

Text #9

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7629. Min Act: -0.1700

Data Index: 372464 (C4 (Web Text))

Max Activating Token Index: 713

Text #10

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 260075 (C4 (Web Text))

Max Activating Token Index: 758

Text #11

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7629. Min Act: -0.1700

Data Index: 1424945 (Python Code)

Max Activating Token Index: 530

Text #12

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 1589707 (Python Code)

Max Activating Token Index: 380

Text #13

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 369016 (C4 (Web Text))

Max Activating Token Index: 384

Text #14

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 1124484 (C4 (Web Text))

Max Activating Token Index: 681

Text #15

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7629. Min Act: -0.1700

Data Index: 41667 (C4 (Web Text))

Max Activating Token Index: 990

Text #16

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 1036021 (C4 (Web Text))

Max Activating Token Index: 462

Text #17

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 1175600 (C4 (Web Text))

Max Activating Token Index: 665

Text #18

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7003. Min Act: -0.1700

Data Index: 1042015 (C4 (Web Text))

Max Activating Token Index: 555

Text #19

Max Range: 3.8880. Min Range: -3.8880

Max Act: 3.7316. Min Act: -0.1700

Data Index: 407070 (C4 (Web Text))

Max Activating Token Index: 659