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

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

Neuron 391 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: 3.5875. Min Range: -3.5875

Max Act: 3.5875. Min Act: -0.1700

Data Index: 1264416 (C4 (Web Text))

Max Activating Token Index: 694

Text #1

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.5248. Min Act: -0.1700

Data Index: 236599 (C4 (Web Text))

Max Activating Token Index: 287

Text #2

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4935. Min Act: -0.1700

Data Index: 298680 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.5248. Min Act: -0.1700

Data Index: 216090 (C4 (Web Text))

Max Activating Token Index: 947

Text #4

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4621. Min Act: -0.1700

Data Index: 204979 (C4 (Web Text))

Max Activating Token Index: 413

Text #5

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4308. Min Act: -0.1700

Data Index: 1006403 (C4 (Web Text))

Max Activating Token Index: 376

Text #6

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4308. Min Act: -0.1700

Data Index: 1365477 (Python Code)

Max Activating Token Index: 385

Text #7

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4308. Min Act: -0.1700

Data Index: 1309132 (C4 (Web Text))

Max Activating Token Index: 389

Text #8

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4621. Min Act: -0.1700

Data Index: 1151704 (C4 (Web Text))

Max Activating Token Index: 478

Text #9

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3837. Min Act: -0.1700

Data Index: 189803 (C4 (Web Text))

Max Activating Token Index: 330

Text #10

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.4308. Min Act: -0.1700

Data Index: 39012 (C4 (Web Text))

Max Activating Token Index: 641

Text #11

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3837. Min Act: -0.1700

Data Index: 345164 (C4 (Web Text))

Max Activating Token Index: 876

Text #12

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3837. Min Act: -0.1700

Data Index: 168879 (C4 (Web Text))

Max Activating Token Index: 737

Text #13

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3994. Min Act: -0.1700

Data Index: 875520 (C4 (Web Text))

Max Activating Token Index: 771

Text #14

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3680. Min Act: -0.1700

Data Index: 635454 (C4 (Web Text))

Max Activating Token Index: 845

Text #15

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3680. Min Act: -0.1700

Data Index: 912923 (C4 (Web Text))

Max Activating Token Index: 177

Text #16

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3680. Min Act: -0.1700

Data Index: 680040 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3523. Min Act: -0.1700

Data Index: 113299 (C4 (Web Text))

Max Activating Token Index: 418

Text #18

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3680. Min Act: -0.1700

Data Index: 1219514 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 3.5875. Min Range: -3.5875

Max Act: 3.3523. Min Act: -0.1700

Data Index: 415608 (C4 (Web Text))

Max Activating Token Index: 792