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

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

Neuron 687 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: 3.2001. Min Range: -3.2001

Max Act: 3.2001. Min Act: -0.1700

Data Index: 952412 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.2001. Min Range: -3.2001

Max Act: 3.0898. Min Act: -0.1700

Data Index: 621662 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.2001. Min Range: -3.2001

Max Act: 3.0267. Min Act: -0.1700

Data Index: 1503642 (Python Code)

Max Activating Token Index: 517

Text #3

Max Range: 3.2001. Min Range: -3.2001

Max Act: 3.0109. Min Act: -0.1700

Data Index: 1558279 (Python Code)

Max Activating Token Index: 572

Text #4

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9951. Min Act: -0.1700

Data Index: 1497063 (Python Code)

Max Activating Token Index: 418

Text #5

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9951. Min Act: -0.1700

Data Index: 1586421 (Python Code)

Max Activating Token Index: 518

Text #6

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9159. Min Act: -0.1700

Data Index: 1444200 (Python Code)

Max Activating Token Index: 538

Text #7

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9793. Min Act: -0.1700

Data Index: 1423827 (Python Code)

Max Activating Token Index: 344

Text #8

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9159. Min Act: -0.1700

Data Index: 1527951 (Python Code)

Max Activating Token Index: 700

Text #9

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9476. Min Act: -0.1700

Data Index: 1578730 (Python Code)

Max Activating Token Index: 722

Text #10

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9159. Min Act: -0.1700

Data Index: 1545213 (Python Code)

Max Activating Token Index: 646

Text #11

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9793. Min Act: -0.1700

Data Index: 1394961 (Python Code)

Max Activating Token Index: 746

Text #12

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9951. Min Act: -0.1700

Data Index: 1654393 (Python Code)

Max Activating Token Index: 608

Text #13

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9001. Min Act: -0.1700

Data Index: 1633794 (Python Code)

Max Activating Token Index: 850

Text #14

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9634. Min Act: -0.1700

Data Index: 1380522 (Python Code)

Max Activating Token Index: 847

Text #15

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9476. Min Act: -0.1700

Data Index: 1396363 (Python Code)

Max Activating Token Index: 196

Text #16

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9793. Min Act: -0.1700

Data Index: 1560792 (Python Code)

Max Activating Token Index: 660

Text #17

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9476. Min Act: -0.1700

Data Index: 1404383 (Python Code)

Max Activating Token Index: 556

Text #18

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9476. Min Act: -0.1700

Data Index: 1524685 (Python Code)

Max Activating Token Index: 992

Text #19

Max Range: 3.2001. Min Range: -3.2001

Max Act: 2.9159. Min Act: -0.1700

Data Index: 1555988 (Python Code)

Max Activating Token Index: 893