Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 981 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.3673. Min Act: -0.1700

Data Index: 1448298 (Python Code)

Max Activating Token Index: 1014

Text #1

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.3045. Min Act: -0.1700

Data Index: 1606413 (Python Code)

Max Activating Token Index: 27

Text #2

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.2573. Min Act: -0.1700

Data Index: 1471877 (Python Code)

Max Activating Token Index: 401

Text #3

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.2416. Min Act: -0.1700

Data Index: 1134008 (C4 (Web Text))

Max Activating Token Index: 342

Text #4

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.2416. Min Act: -0.1700

Data Index: 1485107 (Python Code)

Max Activating Token Index: 621

Text #5

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.2259. Min Act: -0.1700

Data Index: 1543825 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1944. Min Act: -0.1700

Data Index: 1452824 (Python Code)

Max Activating Token Index: 886

Text #7

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1787. Min Act: -0.1700

Data Index: 198557 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1630. Min Act: -0.1700

Data Index: 1621670 (Python Code)

Max Activating Token Index: 167

Text #9

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1787. Min Act: -0.1700

Data Index: 1369735 (Python Code)

Max Activating Token Index: 659

Text #10

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1787. Min Act: -0.1700

Data Index: 1255147 (C4 (Web Text))

Max Activating Token Index: 654

Text #11

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1630. Min Act: -0.1700

Data Index: 1445248 (Python Code)

Max Activating Token Index: 397

Text #12

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1315. Min Act: -0.1700

Data Index: 1552676 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1472. Min Act: -0.1700

Data Index: 1475572 (Python Code)

Max Activating Token Index: 556

Text #14

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1472. Min Act: -0.1700

Data Index: 1149311 (C4 (Web Text))

Max Activating Token Index: 69

Text #15

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1315. Min Act: -0.1700

Data Index: 1514961 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1315. Min Act: -0.1700

Data Index: 1384362 (Python Code)

Max Activating Token Index: 720

Text #17

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1315. Min Act: -0.1700

Data Index: 467310 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1157. Min Act: -0.1700

Data Index: 1597405 (Python Code)

Max Activating Token Index: 738

Text #19

Max Range: 3.3673. Min Range: -3.3673

Max Act: 3.1157. Min Act: -0.1700

Data Index: 1471856 (Python Code)

Max Activating Token Index: 551