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

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

Neuron 1154 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: 3.6916. Min Range: -3.6916

Max Act: 3.6603. Min Act: -0.1700

Data Index: 1040596 (C4 (Web Text))

Max Activating Token Index: 690

Text #1

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.6603. Min Act: -0.1700

Data Index: 1143578 (C4 (Web Text))

Max Activating Token Index: 446

Text #2

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.6916. Min Act: -0.1700

Data Index: 1321574 (C4 (Web Text))

Max Activating Token Index: 1011

Text #3

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.6603. Min Act: -0.1700

Data Index: 1087813 (C4 (Web Text))

Max Activating Token Index: 869

Text #4

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.6290. Min Act: -0.1700

Data Index: 1221831 (C4 (Web Text))

Max Activating Token Index: 370

Text #5

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.6290. Min Act: -0.1700

Data Index: 8919 (C4 (Web Text))

Max Activating Token Index: 844

Text #6

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5664. Min Act: -0.1700

Data Index: 37133 (C4 (Web Text))

Max Activating Token Index: 635

Text #7

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5977. Min Act: -0.1700

Data Index: 1024109 (C4 (Web Text))

Max Activating Token Index: 192

Text #8

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5977. Min Act: -0.1700

Data Index: 1313265 (C4 (Web Text))

Max Activating Token Index: 717

Text #9

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5664. Min Act: -0.1700

Data Index: 1198645 (C4 (Web Text))

Max Activating Token Index: 394

Text #10

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5664. Min Act: -0.1700

Data Index: 1081060 (C4 (Web Text))

Max Activating Token Index: 204

Text #11

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5664. Min Act: -0.1700

Data Index: 377524 (C4 (Web Text))

Max Activating Token Index: 570

Text #12

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5664. Min Act: -0.1700

Data Index: 821410 (C4 (Web Text))

Max Activating Token Index: 100

Text #13

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5351. Min Act: -0.1700

Data Index: 522526 (C4 (Web Text))

Max Activating Token Index: 396

Text #14

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5351. Min Act: -0.1700

Data Index: 501370 (C4 (Web Text))

Max Activating Token Index: 259

Text #15

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5351. Min Act: -0.1700

Data Index: 539200 (C4 (Web Text))

Max Activating Token Index: 745

Text #16

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5351. Min Act: -0.1700

Data Index: 949976 (C4 (Web Text))

Max Activating Token Index: 60

Text #17

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5037. Min Act: -0.1700

Data Index: 739807 (C4 (Web Text))

Max Activating Token Index: 123

Text #18

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5351. Min Act: -0.1700

Data Index: 1010468 (C4 (Web Text))

Max Activating Token Index: 162

Text #19

Max Range: 3.6916. Min Range: -3.6916

Max Act: 3.5037. Min Act: -0.1700

Data Index: 361365 (C4 (Web Text))

Max Activating Token Index: 83