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

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

Neuron 1223 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.8690. Min Act: -0.1700

Data Index: 903509 (C4 (Web Text))

Max Activating Token Index: 330

Text #1

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.8690. Min Act: -0.1700

Data Index: 1244070 (C4 (Web Text))

Max Activating Token Index: 260

Text #2

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.6813. Min Act: -0.1700

Data Index: 1219885 (C4 (Web Text))

Max Activating Token Index: 844

Text #3

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.6813. Min Act: -0.1700

Data Index: 1043937 (C4 (Web Text))

Max Activating Token Index: 336

Text #4

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5874. Min Act: -0.1700

Data Index: 222610 (C4 (Web Text))

Max Activating Token Index: 959

Text #5

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.6500. Min Act: -0.1700

Data Index: 133796 (C4 (Web Text))

Max Activating Token Index: 934

Text #6

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5560. Min Act: -0.1700

Data Index: 796622 (C4 (Web Text))

Max Activating Token Index: 598

Text #7

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5560. Min Act: -0.1700

Data Index: 302918 (C4 (Web Text))

Max Activating Token Index: 1016

Text #8

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5247. Min Act: -0.1700

Data Index: 1360309 (Python Code)

Max Activating Token Index: 296

Text #9

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5560. Min Act: -0.1700

Data Index: 219981 (C4 (Web Text))

Max Activating Token Index: 950

Text #10

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5560. Min Act: -0.1700

Data Index: 403400 (C4 (Web Text))

Max Activating Token Index: 106

Text #11

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5090. Min Act: -0.1700

Data Index: 1168459 (C4 (Web Text))

Max Activating Token Index: 655

Text #12

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5560. Min Act: -0.1700

Data Index: 448658 (C4 (Web Text))

Max Activating Token Index: 723

Text #13

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5247. Min Act: -0.1700

Data Index: 60208 (C4 (Web Text))

Max Activating Token Index: 76

Text #14

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5247. Min Act: -0.1700

Data Index: 1153616 (C4 (Web Text))

Max Activating Token Index: 692

Text #15

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5247. Min Act: -0.1700

Data Index: 638493 (C4 (Web Text))

Max Activating Token Index: 274

Text #16

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.4777. Min Act: -0.1700

Data Index: 436357 (C4 (Web Text))

Max Activating Token Index: 298

Text #17

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.4620. Min Act: -0.1700

Data Index: 840070 (C4 (Web Text))

Max Activating Token Index: 819

Text #18

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.5247. Min Act: -0.1700

Data Index: 349690 (C4 (Web Text))

Max Activating Token Index: 415

Text #19

Max Range: 3.8690. Min Range: -3.8690

Max Act: 3.4620. Min Act: -0.1700

Data Index: 346152 (C4 (Web Text))

Max Activating Token Index: 830