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

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

Neuron 193 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.6618. Min Act: -0.1700

Data Index: 1456088 (Python Code)

Max Activating Token Index: 2

Text #1

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.6305. Min Act: -0.1700

Data Index: 1157001 (C4 (Web Text))

Max Activating Token Index: 243

Text #2

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5992. Min Act: -0.1700

Data Index: 735014 (C4 (Web Text))

Max Activating Token Index: 806

Text #3

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5836. Min Act: -0.1700

Data Index: 60455 (C4 (Web Text))

Max Activating Token Index: 540

Text #4

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5522. Min Act: -0.1700

Data Index: 1612692 (Python Code)

Max Activating Token Index: 164

Text #5

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5522. Min Act: -0.1700

Data Index: 1512358 (Python Code)

Max Activating Token Index: 815

Text #6

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5522. Min Act: -0.1700

Data Index: 1405377 (Python Code)

Max Activating Token Index: 532

Text #7

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5209. Min Act: -0.1700

Data Index: 648838 (C4 (Web Text))

Max Activating Token Index: 676

Text #8

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5366. Min Act: -0.1700

Data Index: 1420958 (Python Code)

Max Activating Token Index: 464

Text #9

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5209. Min Act: -0.1700

Data Index: 1020504 (C4 (Web Text))

Max Activating Token Index: 637

Text #10

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.4896. Min Act: -0.1700

Data Index: 1423365 (Python Code)

Max Activating Token Index: 773

Text #11

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.4896. Min Act: -0.1700

Data Index: 1371800 (Python Code)

Max Activating Token Index: 830

Text #12

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5209. Min Act: -0.1700

Data Index: 1605763 (Python Code)

Max Activating Token Index: 763

Text #13

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5209. Min Act: -0.1700

Data Index: 1652393 (Python Code)

Max Activating Token Index: 26

Text #14

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5209. Min Act: -0.1700

Data Index: 828917 (C4 (Web Text))

Max Activating Token Index: 491

Text #15

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5052. Min Act: -0.1700

Data Index: 1502797 (Python Code)

Max Activating Token Index: 574

Text #16

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.4896. Min Act: -0.1700

Data Index: 1564065 (Python Code)

Max Activating Token Index: 27

Text #17

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5052. Min Act: -0.1700

Data Index: 1638239 (Python Code)

Max Activating Token Index: 13

Text #18

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5209. Min Act: -0.1700

Data Index: 510185 (C4 (Web Text))

Max Activating Token Index: 595

Text #19

Max Range: 3.6618. Min Range: -3.6618

Max Act: 3.5052. Min Act: -0.1700

Data Index: 538445 (C4 (Web Text))

Max Activating Token Index: 459