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

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

Neuron 1620 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: 3.7662. Min Range: -3.7662

Max Act: 3.7662. Min Act: -0.1700

Data Index: 933008 (C4 (Web Text))

Max Activating Token Index: 877

Text #1

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.7036. Min Act: -0.1700

Data Index: 629019 (C4 (Web Text))

Max Activating Token Index: 624

Text #2

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.7036. Min Act: -0.1700

Data Index: 1048147 (C4 (Web Text))

Max Activating Token Index: 79

Text #3

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.7036. Min Act: -0.1700

Data Index: 976012 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.7036. Min Act: -0.1700

Data Index: 908232 (C4 (Web Text))

Max Activating Token Index: 54

Text #5

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 1019168 (C4 (Web Text))

Max Activating Token Index: 905

Text #6

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.6410. Min Act: -0.1700

Data Index: 1044776 (C4 (Web Text))

Max Activating Token Index: 478

Text #7

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 598062 (C4 (Web Text))

Max Activating Token Index: 808

Text #8

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.6097. Min Act: -0.1700

Data Index: 232910 (C4 (Web Text))

Max Activating Token Index: 369

Text #9

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.6410. Min Act: -0.1700

Data Index: 434472 (C4 (Web Text))

Max Activating Token Index: 383

Text #10

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.6097. Min Act: -0.1700

Data Index: 208722 (C4 (Web Text))

Max Activating Token Index: 344

Text #11

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 496065 (C4 (Web Text))

Max Activating Token Index: 528

Text #12

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 986498 (C4 (Web Text))

Max Activating Token Index: 483

Text #13

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5471. Min Act: -0.1700

Data Index: 501713 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 28850 (C4 (Web Text))

Max Activating Token Index: 854

Text #15

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 525229 (C4 (Web Text))

Max Activating Token Index: 350

Text #16

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5471. Min Act: -0.1700

Data Index: 890877 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.6097. Min Act: -0.1700

Data Index: 654887 (C4 (Web Text))

Max Activating Token Index: 453

Text #18

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.5784. Min Act: -0.1700

Data Index: 175547 (C4 (Web Text))

Max Activating Token Index: 459

Text #19

Max Range: 3.7662. Min Range: -3.7662

Max Act: 3.4844. Min Act: -0.1700

Data Index: 976861 (C4 (Web Text))

Max Activating Token Index: 871