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

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

Neuron 1350 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: 3.8249. Min Range: -3.8249

Max Act: 3.8249. Min Act: -0.1700

Data Index: 644007 (C4 (Web Text))

Max Activating Token Index: 157

Text #1

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.7311. Min Act: -0.1700

Data Index: 1164707 (C4 (Web Text))

Max Activating Token Index: 892

Text #2

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.7311. Min Act: -0.1700

Data Index: 167508 (C4 (Web Text))

Max Activating Token Index: 119

Text #3

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.7311. Min Act: -0.1700

Data Index: 17954 (C4 (Web Text))

Max Activating Token Index: 889

Text #4

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6998. Min Act: -0.1700

Data Index: 571481 (C4 (Web Text))

Max Activating Token Index: 780

Text #5

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6372. Min Act: -0.1700

Data Index: 733193 (C4 (Web Text))

Max Activating Token Index: 834

Text #6

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6059. Min Act: -0.1700

Data Index: 69240 (C4 (Web Text))

Max Activating Token Index: 895

Text #7

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6372. Min Act: -0.1700

Data Index: 1183269 (C4 (Web Text))

Max Activating Token Index: 738

Text #8

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6059. Min Act: -0.1700

Data Index: 304592 (C4 (Web Text))

Max Activating Token Index: 704

Text #9

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6372. Min Act: -0.1700

Data Index: 574472 (C4 (Web Text))

Max Activating Token Index: 544

Text #10

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6059. Min Act: -0.1700

Data Index: 76219 (C4 (Web Text))

Max Activating Token Index: 642

Text #11

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5746. Min Act: -0.1700

Data Index: 308937 (C4 (Web Text))

Max Activating Token Index: 939

Text #12

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.6372. Min Act: -0.1700

Data Index: 1248981 (C4 (Web Text))

Max Activating Token Index: 647

Text #13

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5746. Min Act: -0.1700

Data Index: 667595 (C4 (Web Text))

Max Activating Token Index: 221

Text #14

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5432. Min Act: -0.1700

Data Index: 45893 (C4 (Web Text))

Max Activating Token Index: 543

Text #15

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5432. Min Act: -0.1700

Data Index: 1011057 (C4 (Web Text))

Max Activating Token Index: 472

Text #16

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5746. Min Act: -0.1700

Data Index: 549203 (C4 (Web Text))

Max Activating Token Index: 442

Text #17

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5432. Min Act: -0.1700

Data Index: 548950 (C4 (Web Text))

Max Activating Token Index: 667

Text #18

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5432. Min Act: -0.1700

Data Index: 966901 (C4 (Web Text))

Max Activating Token Index: 652

Text #19

Max Range: 3.8249. Min Range: -3.8249

Max Act: 3.5119. Min Act: -0.1700

Data Index: 633913 (C4 (Web Text))

Max Activating Token Index: 176