Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 2007 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.5622. Min Act: -0.1700

Data Index: 1213263 (C4 (Web Text))

Max Activating Token Index: 840

Text #1

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.5308. Min Act: -0.1700

Data Index: 301647 (C4 (Web Text))

Max Activating Token Index: 10

Text #2

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.5308. Min Act: -0.1700

Data Index: 943016 (C4 (Web Text))

Max Activating Token Index: 64

Text #3

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4995. Min Act: -0.1700

Data Index: 1324474 (C4 (Web Text))

Max Activating Token Index: 446

Text #4

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4995. Min Act: -0.1700

Data Index: 117882 (C4 (Web Text))

Max Activating Token Index: 13

Text #5

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4682. Min Act: -0.1700

Data Index: 482676 (C4 (Web Text))

Max Activating Token Index: 825

Text #6

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4682. Min Act: -0.1700

Data Index: 754209 (C4 (Web Text))

Max Activating Token Index: 98

Text #7

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4368. Min Act: -0.1700

Data Index: 1417149 (Python Code)

Max Activating Token Index: 21

Text #8

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 1137289 (C4 (Web Text))

Max Activating Token Index: 402

Text #9

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4368. Min Act: -0.1700

Data Index: 1158425 (C4 (Web Text))

Max Activating Token Index: 77

Text #10

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4368. Min Act: -0.1700

Data Index: 1264493 (C4 (Web Text))

Max Activating Token Index: 91

Text #11

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4368. Min Act: -0.1700

Data Index: 1161069 (C4 (Web Text))

Max Activating Token Index: 260

Text #12

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4368. Min Act: -0.1700

Data Index: 1311108 (C4 (Web Text))

Max Activating Token Index: 13

Text #13

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 775865 (C4 (Web Text))

Max Activating Token Index: 261

Text #14

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 709702 (C4 (Web Text))

Max Activating Token Index: 21

Text #15

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 793317 (C4 (Web Text))

Max Activating Token Index: 411

Text #16

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 908104 (C4 (Web Text))

Max Activating Token Index: 15

Text #17

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 236117 (C4 (Web Text))

Max Activating Token Index: 707

Text #18

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 1345614 (C4 (Web Text))

Max Activating Token Index: 30

Text #19

Max Range: 3.5622. Min Range: -3.5622

Max Act: 3.4054. Min Act: -0.1700

Data Index: 26319 (C4 (Web Text))

Max Activating Token Index: 116