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

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

Neuron 2 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.7270. Min Range: -8.7270

Max Act: 8.7270. Min Act: -0.1700

Data Index: 734225 (C4 (Web Text))

Max Activating Token Index: 445

Text #1

Max Range: 8.7270. Min Range: -8.7270

Max Act: 8.1645. Min Act: -0.1700

Data Index: 1244587 (C4 (Web Text))

Max Activating Token Index: 171

Text #2

Max Range: 8.7270. Min Range: -8.7270

Max Act: 8.1645. Min Act: -0.1700

Data Index: 156589 (C4 (Web Text))

Max Activating Token Index: 147

Text #3

Max Range: 8.7270. Min Range: -8.7270

Max Act: 8.1645. Min Act: -0.1700

Data Index: 239731 (C4 (Web Text))

Max Activating Token Index: 971

Text #4

Max Range: 8.7270. Min Range: -8.7270

Max Act: 8.0707. Min Act: -0.1700

Data Index: 1291380 (C4 (Web Text))

Max Activating Token Index: 369

Text #5

Max Range: 8.7270. Min Range: -8.7270

Max Act: 8.0082. Min Act: -0.1700

Data Index: 1312566 (C4 (Web Text))

Max Activating Token Index: 511

Text #6

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.9770. Min Act: -0.1700

Data Index: 619865 (C4 (Web Text))

Max Activating Token Index: 772

Text #7

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.9770. Min Act: -0.1700

Data Index: 1264262 (C4 (Web Text))

Max Activating Token Index: 555

Text #8

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.8832. Min Act: -0.1700

Data Index: 1031594 (C4 (Web Text))

Max Activating Token Index: 280

Text #9

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.9145. Min Act: -0.1700

Data Index: 1274639 (C4 (Web Text))

Max Activating Token Index: 168

Text #10

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.9145. Min Act: -0.1700

Data Index: 1342140 (C4 (Web Text))

Max Activating Token Index: 219

Text #11

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.8520. Min Act: -0.1700

Data Index: 124940 (C4 (Web Text))

Max Activating Token Index: 58

Text #12

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.8832. Min Act: -0.1699

Data Index: 349824 (C4 (Web Text))

Max Activating Token Index: 10

Text #13

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.8520. Min Act: -0.1700

Data Index: 1040443 (C4 (Web Text))

Max Activating Token Index: 190

Text #14

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.9457. Min Act: -0.1700

Data Index: 1002937 (C4 (Web Text))

Max Activating Token Index: 780

Text #15

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.7895. Min Act: -0.1700

Data Index: 21062 (C4 (Web Text))

Max Activating Token Index: 315

Text #16

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.9145. Min Act: -0.1700

Data Index: 102757 (C4 (Web Text))

Max Activating Token Index: 684

Text #17

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.8832. Min Act: -0.1700

Data Index: 46211 (C4 (Web Text))

Max Activating Token Index: 162

Text #18

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.7895. Min Act: -0.1700

Data Index: 586444 (C4 (Web Text))

Max Activating Token Index: 941

Text #19

Max Range: 8.7270. Min Range: -8.7270

Max Act: 7.7270. Min Act: -0.1700

Data Index: 957780 (C4 (Web Text))

Max Activating Token Index: 737