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

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

Neuron 1964 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.3754. Min Act: -0.1698

Data Index: 1270210 (C4 (Web Text))

Max Activating Token Index: 391

Text #1

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.2817. Min Act: -0.1700

Data Index: 220583 (C4 (Web Text))

Max Activating Token Index: 106

Text #2

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0942. Min Act: -0.1699

Data Index: 1035524 (C4 (Web Text))

Max Activating Token Index: 593

Text #3

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0629. Min Act: -0.1699

Data Index: 415458 (C4 (Web Text))

Max Activating Token Index: 818

Text #4

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.1254. Min Act: -0.1699

Data Index: 345288 (C4 (Web Text))

Max Activating Token Index: 191

Text #5

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0629. Min Act: -0.1700

Data Index: 1299128 (C4 (Web Text))

Max Activating Token Index: 813

Text #6

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0942. Min Act: -0.1700

Data Index: 576336 (C4 (Web Text))

Max Activating Token Index: 503

Text #7

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0004. Min Act: -0.1700

Data Index: 572523 (C4 (Web Text))

Max Activating Token Index: 415

Text #8

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0317. Min Act: -0.1700

Data Index: 410947 (C4 (Web Text))

Max Activating Token Index: 985

Text #9

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0317. Min Act: -0.1700

Data Index: 915209 (C4 (Web Text))

Max Activating Token Index: 750

Text #10

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0317. Min Act: -0.1700

Data Index: 891387 (C4 (Web Text))

Max Activating Token Index: 196

Text #11

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0317. Min Act: -0.1700

Data Index: 604811 (C4 (Web Text))

Max Activating Token Index: 566

Text #12

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0004. Min Act: -0.1700

Data Index: 1289119 (C4 (Web Text))

Max Activating Token Index: 762

Text #13

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0317. Min Act: -0.1700

Data Index: 1231963 (C4 (Web Text))

Max Activating Token Index: 625

Text #14

Max Range: 5.3754. Min Range: -5.3754

Max Act: 4.9379. Min Act: -0.1700

Data Index: 323203 (C4 (Web Text))

Max Activating Token Index: 69

Text #15

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0004. Min Act: -0.1700

Data Index: 1269988 (C4 (Web Text))

Max Activating Token Index: 875

Text #16

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0317. Min Act: -0.1700

Data Index: 1251791 (C4 (Web Text))

Max Activating Token Index: 436

Text #17

Max Range: 5.3754. Min Range: -5.3754

Max Act: 5.0004. Min Act: -0.1700

Data Index: 561644 (C4 (Web Text))

Max Activating Token Index: 114

Text #18

Max Range: 5.3754. Min Range: -5.3754

Max Act: 4.9379. Min Act: -0.1700

Data Index: 1147632 (C4 (Web Text))

Max Activating Token Index: 124

Text #19

Max Range: 5.3754. Min Range: -5.3754

Max Act: 4.9067. Min Act: -0.1700

Data Index: 625675 (C4 (Web Text))

Max Activating Token Index: 333