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

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

Neuron 946 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: 5.7014. Min Range: -5.7014

Max Act: 5.7014. Min Act: -0.1700

Data Index: 1477181 (Python Code)

Max Activating Token Index: 994

Text #1

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.6233. Min Act: -0.1700

Data Index: 1485880 (Python Code)

Max Activating Token Index: 321

Text #2

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.4671. Min Act: -0.1700

Data Index: 938530 (C4 (Web Text))

Max Activating Token Index: 576

Text #3

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.3889. Min Act: -0.1700

Data Index: 1629756 (Python Code)

Max Activating Token Index: 367

Text #4

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.2796. Min Act: -0.1700

Data Index: 1550668 (Python Code)

Max Activating Token Index: 693

Text #5

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.2171. Min Act: -0.1700

Data Index: 1574793 (Python Code)

Max Activating Token Index: 129

Text #6

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.2014. Min Act: -0.1700

Data Index: 1470205 (Python Code)

Max Activating Token Index: 953

Text #7

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.1546. Min Act: -0.1700

Data Index: 1041492 (C4 (Web Text))

Max Activating Token Index: 331

Text #8

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.1702. Min Act: -0.1700

Data Index: 1402636 (Python Code)

Max Activating Token Index: 528

Text #9

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.1546. Min Act: -0.1700

Data Index: 1640817 (Python Code)

Max Activating Token Index: 403

Text #10

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.1233. Min Act: -0.1700

Data Index: 1530522 (Python Code)

Max Activating Token Index: 73

Text #11

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0764. Min Act: -0.1700

Data Index: 1361405 (Python Code)

Max Activating Token Index: 659

Text #12

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.1077. Min Act: -0.1700

Data Index: 1621859 (Python Code)

Max Activating Token Index: 858

Text #13

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0764. Min Act: -0.1700

Data Index: 1591214 (Python Code)

Max Activating Token Index: 868

Text #14

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0608. Min Act: -0.1700

Data Index: 1592843 (Python Code)

Max Activating Token Index: 927

Text #15

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0374. Min Act: -0.1699

Data Index: 1387971 (Python Code)

Max Activating Token Index: 645

Text #16

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0374. Min Act: -0.1700

Data Index: 1436043 (Python Code)

Max Activating Token Index: 825

Text #17

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0764. Min Act: -0.1700

Data Index: 1313987 (C4 (Web Text))

Max Activating Token Index: 438

Text #18

Max Range: 5.7014. Min Range: -5.7014

Max Act: 5.0452. Min Act: -0.1700

Data Index: 1627487 (Python Code)

Max Activating Token Index: 360

Text #19

Max Range: 5.7014. Min Range: -5.7014

Max Act: 4.9905. Min Act: -0.1700

Data Index: 163955 (C4 (Web Text))

Max Activating Token Index: 89