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

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

Neuron 122 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: 7.9012. Min Range: -7.9012

Max Act: 7.9012. Min Act: -0.1700

Data Index: 981989 (C4 (Web Text))

Max Activating Token Index: 843

Text #1

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.7762. Min Act: -0.1700

Data Index: 248925 (C4 (Web Text))

Max Activating Token Index: 648

Text #2

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.8387. Min Act: -0.1700

Data Index: 1483101 (Python Code)

Max Activating Token Index: 846

Text #3

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.8387. Min Act: -0.1700

Data Index: 913239 (C4 (Web Text))

Max Activating Token Index: 897

Text #4

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.6199. Min Act: -0.1700

Data Index: 1147656 (C4 (Web Text))

Max Activating Token Index: 767

Text #5

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.6824. Min Act: -0.1700

Data Index: 1612302 (Python Code)

Max Activating Token Index: 687

Text #6

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.5262. Min Act: -0.1700

Data Index: 649416 (C4 (Web Text))

Max Activating Token Index: 147

Text #7

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.7137. Min Act: -0.1700

Data Index: 1527268 (Python Code)

Max Activating Token Index: 944

Text #8

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.5887. Min Act: -0.1700

Data Index: 350755 (C4 (Web Text))

Max Activating Token Index: 668

Text #9

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.6512. Min Act: -0.1700

Data Index: 1487490 (Python Code)

Max Activating Token Index: 893

Text #10

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.6512. Min Act: -0.1700

Data Index: 1629824 (Python Code)

Max Activating Token Index: 825

Text #11

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.5574. Min Act: -0.1700

Data Index: 1654079 (Python Code)

Max Activating Token Index: 731

Text #12

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.6199. Min Act: -0.1700

Data Index: 464214 (C4 (Web Text))

Max Activating Token Index: 378

Text #13

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.6199. Min Act: -0.1700

Data Index: 1377300 (Python Code)

Max Activating Token Index: 826

Text #14

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.5574. Min Act: -0.1700

Data Index: 1647367 (Python Code)

Max Activating Token Index: 924

Text #15

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.4949. Min Act: -0.1700

Data Index: 451938 (C4 (Web Text))

Max Activating Token Index: 250

Text #16

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.4637. Min Act: -0.1700

Data Index: 1076404 (C4 (Web Text))

Max Activating Token Index: 781

Text #17

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.4637. Min Act: -0.1700

Data Index: 525534 (C4 (Web Text))

Max Activating Token Index: 950

Text #18

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.5574. Min Act: -0.1700

Data Index: 743720 (C4 (Web Text))

Max Activating Token Index: 782

Text #19

Max Range: 7.9012. Min Range: -7.9012

Max Act: 7.4949. Min Act: -0.1700

Data Index: 1578665 (Python Code)

Max Activating Token Index: 628