Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 504 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.9559. Min Act: -0.1700

Data Index: 1361542 (Python Code)

Max Activating Token Index: 684

Text #1

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.9559. Min Act: -0.1700

Data Index: 1434978 (Python Code)

Max Activating Token Index: 935

Text #2

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.9559. Min Act: -0.1700

Data Index: 1418006 (Python Code)

Max Activating Token Index: 383

Text #3

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.8621. Min Act: -0.1700

Data Index: 1473289 (Python Code)

Max Activating Token Index: 960

Text #4

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.8308. Min Act: -0.1700

Data Index: 1366685 (Python Code)

Max Activating Token Index: 357

Text #5

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.7996. Min Act: -0.1700

Data Index: 1629712 (Python Code)

Max Activating Token Index: 81

Text #6

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.7057. Min Act: -0.1700

Data Index: 394690 (C4 (Web Text))

Max Activating Token Index: 631

Text #7

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.7057. Min Act: -0.1700

Data Index: 1473637 (Python Code)

Max Activating Token Index: 542

Text #8

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.7370. Min Act: -0.1700

Data Index: 1489911 (Python Code)

Max Activating Token Index: 533

Text #9

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6744. Min Act: -0.1700

Data Index: 1449135 (Python Code)

Max Activating Token Index: 843

Text #10

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6744. Min Act: -0.1700

Data Index: 1439445 (Python Code)

Max Activating Token Index: 581

Text #11

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6744. Min Act: -0.1700

Data Index: 1436463 (Python Code)

Max Activating Token Index: 771

Text #12

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6118. Min Act: -0.1700

Data Index: 1413971 (Python Code)

Max Activating Token Index: 518

Text #13

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6431. Min Act: -0.1700

Data Index: 1649777 (Python Code)

Max Activating Token Index: 536

Text #14

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.5805. Min Act: -0.1700

Data Index: 1524456 (Python Code)

Max Activating Token Index: 410

Text #15

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6118. Min Act: -0.1699

Data Index: 1374286 (Python Code)

Max Activating Token Index: 790

Text #16

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.5805. Min Act: -0.1700

Data Index: 1430678 (Python Code)

Max Activating Token Index: 795

Text #17

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.6118. Min Act: -0.1700

Data Index: 221894 (C4 (Web Text))

Max Activating Token Index: 125

Text #18

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.5805. Min Act: -0.1700

Data Index: 1454362 (Python Code)

Max Activating Token Index: 1000

Text #19

Max Range: 3.9559. Min Range: -3.9559

Max Act: 3.5805. Min Act: -0.1700

Data Index: 1579027 (Python Code)

Max Activating Token Index: 757