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

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

Neuron 1991 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.9568. Min Act: -0.1700

Data Index: 550657 (C4 (Web Text))

Max Activating Token Index: 93

Text #1

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.9256. Min Act: -0.1700

Data Index: 1370512 (Python Code)

Max Activating Token Index: 869

Text #2

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.9256. Min Act: -0.1700

Data Index: 1580556 (Python Code)

Max Activating Token Index: 991

Text #3

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.8630. Min Act: -0.1700

Data Index: 1446539 (Python Code)

Max Activating Token Index: 170

Text #4

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.8630. Min Act: -0.1700

Data Index: 1359897 (Python Code)

Max Activating Token Index: 500

Text #5

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.8005. Min Act: -0.1700

Data Index: 1163131 (C4 (Web Text))

Max Activating Token Index: 934

Text #6

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7692. Min Act: -0.1700

Data Index: 1435792 (Python Code)

Max Activating Token Index: 751

Text #7

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7379. Min Act: -0.1700

Data Index: 916384 (C4 (Web Text))

Max Activating Token Index: 469

Text #8

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7692. Min Act: -0.1700

Data Index: 48060 (C4 (Web Text))

Max Activating Token Index: 599

Text #9

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7692. Min Act: -0.1700

Data Index: 253728 (C4 (Web Text))

Max Activating Token Index: 713

Text #10

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7692. Min Act: -0.1700

Data Index: 1189246 (C4 (Web Text))

Max Activating Token Index: 520

Text #11

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7066. Min Act: -0.1700

Data Index: 177566 (C4 (Web Text))

Max Activating Token Index: 521

Text #12

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7066. Min Act: -0.1700

Data Index: 1561411 (Python Code)

Max Activating Token Index: 710

Text #13

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7379. Min Act: -0.1700

Data Index: 1441321 (Python Code)

Max Activating Token Index: 5

Text #14

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7379. Min Act: -0.1700

Data Index: 1526595 (Python Code)

Max Activating Token Index: 773

Text #15

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7692. Min Act: -0.1700

Data Index: 1515269 (Python Code)

Max Activating Token Index: 344

Text #16

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7379. Min Act: -0.1700

Data Index: 727009 (C4 (Web Text))

Max Activating Token Index: 967

Text #17

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7379. Min Act: -0.1700

Data Index: 1561897 (Python Code)

Max Activating Token Index: 994

Text #18

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7066. Min Act: -0.1700

Data Index: 1364470 (Python Code)

Max Activating Token Index: 13

Text #19

Max Range: 3.9568. Min Range: -3.9568

Max Act: 3.7066. Min Act: -0.1700

Data Index: 1630695 (Python Code)

Max Activating Token Index: 469