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

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

Neuron 1759 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: 5.3997. Min Range: -5.3997

Max Act: 5.3997. Min Act: -0.1700

Data Index: 1394644 (Python Code)

Max Activating Token Index: 747

Text #1

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.3997. Min Act: -0.1700

Data Index: 1407291 (Python Code)

Max Activating Token Index: 937

Text #2

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.3997. Min Act: -0.1700

Data Index: 1538974 (Python Code)

Max Activating Token Index: 686

Text #3

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.3372. Min Act: -0.1700

Data Index: 372563 (C4 (Web Text))

Max Activating Token Index: 102

Text #4

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.3060. Min Act: -0.1700

Data Index: 976860 (C4 (Web Text))

Max Activating Token Index: 451

Text #5

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2747. Min Act: -0.1700

Data Index: 347441 (C4 (Web Text))

Max Activating Token Index: 98

Text #6

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2122. Min Act: -0.1700

Data Index: 970972 (C4 (Web Text))

Max Activating Token Index: 495

Text #7

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2747. Min Act: -0.1700

Data Index: 444702 (C4 (Web Text))

Max Activating Token Index: 65

Text #8

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2435. Min Act: -0.1700

Data Index: 1278026 (C4 (Web Text))

Max Activating Token Index: 446

Text #9

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2435. Min Act: -0.1700

Data Index: 1374599 (Python Code)

Max Activating Token Index: 217

Text #10

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2435. Min Act: -0.1700

Data Index: 236540 (C4 (Web Text))

Max Activating Token Index: 482

Text #11

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2122. Min Act: -0.1700

Data Index: 21617 (C4 (Web Text))

Max Activating Token Index: 711

Text #12

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2122. Min Act: -0.1700

Data Index: 336201 (C4 (Web Text))

Max Activating Token Index: 95

Text #13

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.1497. Min Act: -0.1700

Data Index: 1143712 (C4 (Web Text))

Max Activating Token Index: 678

Text #14

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.1810. Min Act: -0.1700

Data Index: 1113365 (C4 (Web Text))

Max Activating Token Index: 457

Text #15

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.1497. Min Act: -0.1700

Data Index: 910160 (C4 (Web Text))

Max Activating Token Index: 234

Text #16

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.1810. Min Act: -0.1700

Data Index: 1368368 (Python Code)

Max Activating Token Index: 711

Text #17

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.1810. Min Act: -0.1700

Data Index: 842940 (C4 (Web Text))

Max Activating Token Index: 497

Text #18

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.2122. Min Act: -0.1700

Data Index: 91926 (C4 (Web Text))

Max Activating Token Index: 896

Text #19

Max Range: 5.3997. Min Range: -5.3997

Max Act: 5.1497. Min Act: -0.1700

Data Index: 1539327 (Python Code)

Max Activating Token Index: 437