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

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

Neuron 1733 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.4503. Min Range: -7.4503

Max Act: 7.4503. Min Act: -0.1700

Data Index: 1500882 (Python Code)

Max Activating Token Index: 723

Text #1

Max Range: 7.4503. Min Range: -7.4503

Max Act: 7.3253. Min Act: -0.1700

Data Index: 1542389 (Python Code)

Max Activating Token Index: 787

Text #2

Max Range: 7.4503. Min Range: -7.4503

Max Act: 7.0441. Min Act: -0.1700

Data Index: 1386212 (Python Code)

Max Activating Token Index: 977

Text #3

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.7628. Min Act: -0.1700

Data Index: 1637716 (Python Code)

Max Activating Token Index: 808

Text #4

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.7628. Min Act: -0.1700

Data Index: 1555634 (Python Code)

Max Activating Token Index: 749

Text #5

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.7941. Min Act: -0.1700

Data Index: 1450076 (Python Code)

Max Activating Token Index: 834

Text #6

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.7316. Min Act: -0.1700

Data Index: 1602447 (Python Code)

Max Activating Token Index: 455

Text #7

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.7003. Min Act: -0.1700

Data Index: 1586914 (Python Code)

Max Activating Token Index: 577

Text #8

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.5128. Min Act: -0.1700

Data Index: 1514107 (Python Code)

Max Activating Token Index: 970

Text #9

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.6378. Min Act: -0.1700

Data Index: 1561675 (Python Code)

Max Activating Token Index: 860

Text #10

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.6691. Min Act: -0.1700

Data Index: 1548326 (Python Code)

Max Activating Token Index: 306

Text #11

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.6691. Min Act: -0.1700

Data Index: 1632504 (Python Code)

Max Activating Token Index: 563

Text #12

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.6378. Min Act: -0.1700

Data Index: 1614858 (Python Code)

Max Activating Token Index: 599

Text #13

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.5753. Min Act: -0.1700

Data Index: 1556326 (Python Code)

Max Activating Token Index: 824

Text #14

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.4503. Min Act: -0.1700

Data Index: 1636003 (Python Code)

Max Activating Token Index: 817

Text #15

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.5441. Min Act: -0.1700

Data Index: 1623950 (Python Code)

Max Activating Token Index: 994

Text #16

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.5441. Min Act: -0.1700

Data Index: 1578951 (Python Code)

Max Activating Token Index: 677

Text #17

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.5753. Min Act: -0.1700

Data Index: 1594952 (Python Code)

Max Activating Token Index: 811

Text #18

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.5441. Min Act: -0.1700

Data Index: 1408131 (Python Code)

Max Activating Token Index: 736

Text #19

Max Range: 7.4503. Min Range: -7.4503

Max Act: 6.6378. Min Act: -0.1700

Data Index: 1433660 (Python Code)

Max Activating Token Index: 678