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

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

Neuron 337 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: 3.9075. Min Range: -3.9075

Max Act: 3.8762. Min Act: -0.1700

Data Index: 1350536 (C4 (Web Text))

Max Activating Token Index: 174

Text #1

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.9075. Min Act: -0.1700

Data Index: 34286 (C4 (Web Text))

Max Activating Token Index: 766

Text #2

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.8762. Min Act: -0.1700

Data Index: 689546 (C4 (Web Text))

Max Activating Token Index: 575

Text #3

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7823. Min Act: -0.1700

Data Index: 1078070 (C4 (Web Text))

Max Activating Token Index: 72

Text #4

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7823. Min Act: -0.1700

Data Index: 1131289 (C4 (Web Text))

Max Activating Token Index: 296

Text #5

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7823. Min Act: -0.1700

Data Index: 1000715 (C4 (Web Text))

Max Activating Token Index: 114

Text #6

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7511. Min Act: -0.1700

Data Index: 6225 (C4 (Web Text))

Max Activating Token Index: 189

Text #7

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6885. Min Act: -0.1700

Data Index: 659950 (C4 (Web Text))

Max Activating Token Index: 166

Text #8

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7511. Min Act: -0.1700

Data Index: 1101201 (C4 (Web Text))

Max Activating Token Index: 61

Text #9

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7198. Min Act: -0.1700

Data Index: 2984 (C4 (Web Text))

Max Activating Token Index: 550

Text #10

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.7198. Min Act: -0.1700

Data Index: 855642 (C4 (Web Text))

Max Activating Token Index: 604

Text #11

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6885. Min Act: -0.1700

Data Index: 674425 (C4 (Web Text))

Max Activating Token Index: 992

Text #12

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6885. Min Act: -0.1700

Data Index: 1352363 (C4 (Web Text))

Max Activating Token Index: 74

Text #13

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6259. Min Act: -0.1700

Data Index: 725248 (C4 (Web Text))

Max Activating Token Index: 142

Text #14

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6885. Min Act: -0.1700

Data Index: 280206 (C4 (Web Text))

Max Activating Token Index: 602

Text #15

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6572. Min Act: -0.1700

Data Index: 588648 (C4 (Web Text))

Max Activating Token Index: 915

Text #16

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.6885. Min Act: -0.1700

Data Index: 133378 (C4 (Web Text))

Max Activating Token Index: 832

Text #17

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.5946. Min Act: -0.1700

Data Index: 315904 (C4 (Web Text))

Max Activating Token Index: 302

Text #18

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.5946. Min Act: -0.1700

Data Index: 280587 (C4 (Web Text))

Max Activating Token Index: 778

Text #19

Max Range: 3.9075. Min Range: -3.9075

Max Act: 3.5946. Min Act: -0.1700

Data Index: 739469 (C4 (Web Text))

Max Activating Token Index: 398