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

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

Neuron 1574 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: 3.6892. Min Range: -3.6892

Max Act: 3.6892. Min Act: -0.1700

Data Index: 1421058 (Python Code)

Max Activating Token Index: 451

Text #1

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5640. Min Act: -0.1700

Data Index: 1521272 (Python Code)

Max Activating Token Index: 679

Text #2

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5327. Min Act: -0.1700

Data Index: 721769 (C4 (Web Text))

Max Activating Token Index: 426

Text #3

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5170. Min Act: -0.1700

Data Index: 1495637 (Python Code)

Max Activating Token Index: 777

Text #4

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5327. Min Act: -0.1700

Data Index: 1541589 (Python Code)

Max Activating Token Index: 822

Text #5

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5640. Min Act: -0.1700

Data Index: 1424373 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5327. Min Act: -0.1700

Data Index: 1365629 (Python Code)

Max Activating Token Index: 593

Text #7

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5014. Min Act: -0.1700

Data Index: 1409580 (Python Code)

Max Activating Token Index: 590

Text #8

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5014. Min Act: -0.1700

Data Index: 1423194 (Python Code)

Max Activating Token Index: 810

Text #9

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5014. Min Act: -0.1700

Data Index: 1454914 (Python Code)

Max Activating Token Index: 826

Text #10

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4857. Min Act: -0.1700

Data Index: 1552265 (Python Code)

Max Activating Token Index: 612

Text #11

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5014. Min Act: -0.1700

Data Index: 1505516 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5014. Min Act: -0.1700

Data Index: 1008533 (C4 (Web Text))

Max Activating Token Index: 899

Text #13

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4543. Min Act: -0.1700

Data Index: 1424773 (Python Code)

Max Activating Token Index: 681

Text #14

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.5014. Min Act: -0.1700

Data Index: 1627973 (Python Code)

Max Activating Token Index: 655

Text #15

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4700. Min Act: -0.1700

Data Index: 1366795 (Python Code)

Max Activating Token Index: 815

Text #16

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4543. Min Act: -0.1700

Data Index: 1367135 (Python Code)

Max Activating Token Index: 575

Text #17

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4857. Min Act: -0.1700

Data Index: 1475171 (Python Code)

Max Activating Token Index: 348

Text #18

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4700. Min Act: -0.1700

Data Index: 1368519 (Python Code)

Max Activating Token Index: 101

Text #19

Max Range: 3.6892. Min Range: -3.6892

Max Act: 3.4543. Min Act: -0.1700

Data Index: 1361006 (Python Code)

Max Activating Token Index: 887