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

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

Neuron 1932 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: 3.7171. Min Range: -3.7171

Max Act: 3.7171. Min Act: -0.1699

Data Index: 1499763 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.6545. Min Act: -0.1700

Data Index: 1553580 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.5606. Min Act: -0.1700

Data Index: 1567303 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.4979. Min Act: -0.1700

Data Index: 1590869 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.4665. Min Act: -0.1700

Data Index: 1568557 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.4352. Min Act: -0.1694

Data Index: 424465 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.4352. Min Act: -0.1694

Data Index: 1252030 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.4038. Min Act: -0.1700

Data Index: 1200421 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.4038. Min Act: -0.1700

Data Index: 1618264 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3568. Min Act: -0.1700

Data Index: 737770 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3254. Min Act: -0.1681

Data Index: 1184985 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3254. Min Act: -0.1690

Data Index: 618797 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3411. Min Act: -0.1700

Data Index: 1525411 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3411. Min Act: -0.1700

Data Index: 1450049 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3097. Min Act: -0.1700

Data Index: 950011 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3097. Min Act: -0.1700

Data Index: 1481073 (Python Code)

Max Activating Token Index: 762

Text #16

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.3254. Min Act: -0.1698

Data Index: 1469587 (Python Code)

Max Activating Token Index: 533

Text #17

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.2940. Min Act: -0.1700

Data Index: 1386618 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.2940. Min Act: -0.1699

Data Index: 1526890 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 3.7171. Min Range: -3.7171

Max Act: 3.2940. Min Act: -0.1700

Data Index: 1625728 (Python Code)

Max Activating Token Index: 1023