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

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

Neuron 979 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.1305. Min Range: -8.1305

Max Act: 8.1305. Min Act: -0.1700

Data Index: 1568186 (Python Code)

Max Activating Token Index: 885

Text #1

Max Range: 8.1305. Min Range: -8.1305

Max Act: 8.0993. Min Act: -0.1700

Data Index: 1613518 (Python Code)

Max Activating Token Index: 996

Text #2

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.8180. Min Act: -0.1700

Data Index: 1522993 (Python Code)

Max Activating Token Index: 849

Text #3

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.7555. Min Act: -0.1700

Data Index: 1298846 (C4 (Web Text))

Max Activating Token Index: 305

Text #4

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.8493. Min Act: -0.1700

Data Index: 1424658 (Python Code)

Max Activating Token Index: 428

Text #5

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.7868. Min Act: -0.1700

Data Index: 1447797 (Python Code)

Max Activating Token Index: 679

Text #6

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.7555. Min Act: -0.1700

Data Index: 1501535 (Python Code)

Max Activating Token Index: 277

Text #7

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.7555. Min Act: -0.1700

Data Index: 1418426 (Python Code)

Max Activating Token Index: 434

Text #8

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6930. Min Act: -0.1700

Data Index: 454379 (C4 (Web Text))

Max Activating Token Index: 806

Text #9

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6930. Min Act: -0.1700

Data Index: 1451373 (Python Code)

Max Activating Token Index: 204

Text #10

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6618. Min Act: -0.1700

Data Index: 1407709 (Python Code)

Max Activating Token Index: 635

Text #11

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.7555. Min Act: -0.1696

Data Index: 1450851 (Python Code)

Max Activating Token Index: 144

Text #12

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6930. Min Act: -0.1700

Data Index: 1443335 (Python Code)

Max Activating Token Index: 765

Text #13

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6305. Min Act: -0.1700

Data Index: 1625221 (Python Code)

Max Activating Token Index: 375

Text #14

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6618. Min Act: -0.1700

Data Index: 1590727 (Python Code)

Max Activating Token Index: 393

Text #15

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6305. Min Act: -0.1700

Data Index: 1420689 (Python Code)

Max Activating Token Index: 159

Text #16

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.5993. Min Act: -0.1700

Data Index: 1575566 (Python Code)

Max Activating Token Index: 419

Text #17

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6305. Min Act: -0.1700

Data Index: 1446252 (Python Code)

Max Activating Token Index: 834

Text #18

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.6930. Min Act: -0.1700

Data Index: 1564906 (Python Code)

Max Activating Token Index: 902

Text #19

Max Range: 8.1305. Min Range: -8.1305

Max Act: 7.5680. Min Act: -0.1700

Data Index: 1341624 (C4 (Web Text))

Max Activating Token Index: 755