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

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

Neuron 674 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.8527. Min Act: -0.1700

Data Index: 907405 (C4 (Web Text))

Max Activating Token Index: 783

Text #1

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.5402. Min Act: -0.1700

Data Index: 109055 (C4 (Web Text))

Max Activating Token Index: 701

Text #2

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.4152. Min Act: -0.1700

Data Index: 199396 (C4 (Web Text))

Max Activating Token Index: 158

Text #3

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.4777. Min Act: -0.1699

Data Index: 1123602 (C4 (Web Text))

Max Activating Token Index: 146

Text #4

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1964. Min Act: -0.1700

Data Index: 682393 (C4 (Web Text))

Max Activating Token Index: 117

Text #5

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.2589. Min Act: -0.1700

Data Index: 142690 (C4 (Web Text))

Max Activating Token Index: 97

Text #6

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1652. Min Act: -0.1700

Data Index: 794286 (C4 (Web Text))

Max Activating Token Index: 415

Text #7

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1652. Min Act: -0.1700

Data Index: 609945 (C4 (Web Text))

Max Activating Token Index: 86

Text #8

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1339. Min Act: -0.1700

Data Index: 842267 (C4 (Web Text))

Max Activating Token Index: 204

Text #9

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1652. Min Act: -0.1700

Data Index: 515734 (C4 (Web Text))

Max Activating Token Index: 127

Text #10

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1339. Min Act: -0.1700

Data Index: 93251 (C4 (Web Text))

Max Activating Token Index: 154

Text #11

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1339. Min Act: -0.1700

Data Index: 109177 (C4 (Web Text))

Max Activating Token Index: 221

Text #12

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1027. Min Act: -0.1700

Data Index: 535908 (C4 (Web Text))

Max Activating Token Index: 686

Text #13

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1027. Min Act: -0.1700

Data Index: 1006501 (C4 (Web Text))

Max Activating Token Index: 158

Text #14

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1339. Min Act: -0.1700

Data Index: 1218324 (C4 (Web Text))

Max Activating Token Index: 477

Text #15

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1339. Min Act: -0.1700

Data Index: 1214449 (C4 (Web Text))

Max Activating Token Index: 82

Text #16

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1652. Min Act: -0.1700

Data Index: 537027 (C4 (Web Text))

Max Activating Token Index: 215

Text #17

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.0714. Min Act: -0.1700

Data Index: 252896 (C4 (Web Text))

Max Activating Token Index: 49

Text #18

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1027. Min Act: -0.1700

Data Index: 710850 (C4 (Web Text))

Max Activating Token Index: 181

Text #19

Max Range: 5.8527. Min Range: -5.8527

Max Act: 5.1027. Min Act: -0.1700

Data Index: 822125 (C4 (Web Text))

Max Activating Token Index: 197