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

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

Neuron 2040 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: 3.6264. Min Range: -3.6264

Max Act: 3.6264. Min Act: -0.1700

Data Index: 960510 (C4 (Web Text))

Max Activating Token Index: 646

Text #1

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5638. Min Act: -0.1700

Data Index: 1106350 (C4 (Web Text))

Max Activating Token Index: 713

Text #2

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5638. Min Act: -0.1700

Data Index: 829886 (C4 (Web Text))

Max Activating Token Index: 712

Text #3

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5324. Min Act: -0.1700

Data Index: 638862 (C4 (Web Text))

Max Activating Token Index: 483

Text #4

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5638. Min Act: -0.1700

Data Index: 374945 (C4 (Web Text))

Max Activating Token Index: 838

Text #5

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5324. Min Act: -0.1700

Data Index: 14689 (C4 (Web Text))

Max Activating Token Index: 581

Text #6

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5324. Min Act: -0.1700

Data Index: 1112911 (C4 (Web Text))

Max Activating Token Index: 630

Text #7

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 930918 (C4 (Web Text))

Max Activating Token Index: 765

Text #8

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 1224061 (C4 (Web Text))

Max Activating Token Index: 167

Text #9

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 534655 (C4 (Web Text))

Max Activating Token Index: 483

Text #10

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 108972 (C4 (Web Text))

Max Activating Token Index: 701

Text #11

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.4698. Min Act: -0.1700

Data Index: 1188838 (C4 (Web Text))

Max Activating Token Index: 321

Text #12

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 51357 (C4 (Web Text))

Max Activating Token Index: 795

Text #13

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5324. Min Act: -0.1700

Data Index: 1028135 (C4 (Web Text))

Max Activating Token Index: 842

Text #14

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.4698. Min Act: -0.1700

Data Index: 188663 (C4 (Web Text))

Max Activating Token Index: 48

Text #15

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 199685 (C4 (Web Text))

Max Activating Token Index: 97

Text #16

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.5011. Min Act: -0.1700

Data Index: 675113 (C4 (Web Text))

Max Activating Token Index: 321

Text #17

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.4384. Min Act: -0.1700

Data Index: 89553 (C4 (Web Text))

Max Activating Token Index: 274

Text #18

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.4384. Min Act: -0.1700

Data Index: 930914 (C4 (Web Text))

Max Activating Token Index: 1021

Text #19

Max Range: 3.6264. Min Range: -3.6264

Max Act: 3.4698. Min Act: -0.1700

Data Index: 244454 (C4 (Web Text))

Max Activating Token Index: 715