Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 92 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.9212. Min Act: -0.1599

Data Index: 389895 (C4 (Web Text))

Max Activating Token Index: 313

Text #1

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.7336. Min Act: -0.1700

Data Index: 778248 (C4 (Web Text))

Max Activating Token Index: 694

Text #2

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.6710. Min Act: -0.1698

Data Index: 664798 (C4 (Web Text))

Max Activating Token Index: 528

Text #3

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.6710. Min Act: -0.1700

Data Index: 953869 (C4 (Web Text))

Max Activating Token Index: 271

Text #4

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.6084. Min Act: -0.1700

Data Index: 484266 (C4 (Web Text))

Max Activating Token Index: 485

Text #5

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5770. Min Act: -0.1448

Data Index: 626486 (C4 (Web Text))

Max Activating Token Index: 604

Text #6

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5457. Min Act: -0.1653

Data Index: 1300598 (C4 (Web Text))

Max Activating Token Index: 660

Text #7

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5770. Min Act: -0.0565

Data Index: 1221619 (C4 (Web Text))

Max Activating Token Index: 720

Text #8

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5770. Min Act: -0.1617

Data Index: 964681 (C4 (Web Text))

Max Activating Token Index: 257

Text #9

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5457. Min Act: -0.1118

Data Index: 109084 (C4 (Web Text))

Max Activating Token Index: 962

Text #10

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5457. Min Act: -0.1679

Data Index: 266512 (C4 (Web Text))

Max Activating Token Index: 890

Text #11

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5457. Min Act: -0.1672

Data Index: 848262 (C4 (Web Text))

Max Activating Token Index: 739

Text #12

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1700

Data Index: 1433596 (Python Code)

Max Activating Token Index: 676

Text #13

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1700

Data Index: 9623 (C4 (Web Text))

Max Activating Token Index: 155

Text #14

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5457. Min Act: -0.1700

Data Index: 316441 (C4 (Web Text))

Max Activating Token Index: 952

Text #15

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1406

Data Index: 114213 (C4 (Web Text))

Max Activating Token Index: 860

Text #16

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1699

Data Index: 904518 (C4 (Web Text))

Max Activating Token Index: 960

Text #17

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1700

Data Index: 26560 (C4 (Web Text))

Max Activating Token Index: 901

Text #18

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1699

Data Index: 1173388 (C4 (Web Text))

Max Activating Token Index: 104

Text #19

Max Range: 3.9212. Min Range: -3.9212

Max Act: 3.5144. Min Act: -0.1695

Data Index: 1266474 (C4 (Web Text))

Max Activating Token Index: 615