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

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

Neuron 726 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: 5.7478. Min Range: -5.7478

Max Act: 5.7478. Min Act: -0.1700

Data Index: 1440020 (Python Code)

Max Activating Token Index: 863

Text #1

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.7478. Min Act: -0.1700

Data Index: 1331901 (C4 (Web Text))

Max Activating Token Index: 285

Text #2

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.6853. Min Act: -0.1700

Data Index: 1516393 (Python Code)

Max Activating Token Index: 777

Text #3

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.6541. Min Act: -0.1700

Data Index: 1644428 (Python Code)

Max Activating Token Index: 853

Text #4

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5916. Min Act: -0.1700

Data Index: 1491811 (Python Code)

Max Activating Token Index: 993

Text #5

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.6228. Min Act: -0.1700

Data Index: 1436181 (Python Code)

Max Activating Token Index: 923

Text #6

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.6228. Min Act: -0.1700

Data Index: 1488202 (Python Code)

Max Activating Token Index: 991

Text #7

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5916. Min Act: -0.1700

Data Index: 1471028 (Python Code)

Max Activating Token Index: 676

Text #8

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5603. Min Act: -0.1700

Data Index: 1391269 (Python Code)

Max Activating Token Index: 980

Text #9

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5603. Min Act: -0.1700

Data Index: 1502945 (Python Code)

Max Activating Token Index: 784

Text #10

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.4666. Min Act: -0.1700

Data Index: 1438897 (Python Code)

Max Activating Token Index: 876

Text #11

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5603. Min Act: -0.1700

Data Index: 1490793 (Python Code)

Max Activating Token Index: 786

Text #12

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5291. Min Act: -0.1700

Data Index: 1388826 (Python Code)

Max Activating Token Index: 208

Text #13

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.5291. Min Act: -0.1700

Data Index: 1615700 (Python Code)

Max Activating Token Index: 920

Text #14

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.4978. Min Act: -0.1700

Data Index: 1469820 (Python Code)

Max Activating Token Index: 517

Text #15

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.4353. Min Act: -0.1700

Data Index: 1558765 (Python Code)

Max Activating Token Index: 158

Text #16

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.4041. Min Act: -0.1700

Data Index: 296805 (C4 (Web Text))

Max Activating Token Index: 900

Text #17

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.3416. Min Act: -0.1700

Data Index: 1433547 (Python Code)

Max Activating Token Index: 875

Text #18

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.4666. Min Act: -0.1700

Data Index: 757329 (C4 (Web Text))

Max Activating Token Index: 762

Text #19

Max Range: 5.7478. Min Range: -5.7478

Max Act: 5.4041. Min Act: -0.1700

Data Index: 1583086 (Python Code)

Max Activating Token Index: 930