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

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

Neuron 458 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: 5.2810. Min Range: -5.2810

Max Act: 5.2810. Min Act: -0.1700

Data Index: 1552468 (Python Code)

Max Activating Token Index: 791

Text #1

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.2498. Min Act: -0.1700

Data Index: 1371948 (Python Code)

Max Activating Token Index: 676

Text #2

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.1873. Min Act: -0.1700

Data Index: 1614874 (Python Code)

Max Activating Token Index: 922

Text #3

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.1560. Min Act: -0.1700

Data Index: 1432002 (Python Code)

Max Activating Token Index: 21

Text #4

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.1560. Min Act: -0.1700

Data Index: 1386047 (Python Code)

Max Activating Token Index: 42

Text #5

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.1248. Min Act: -0.1700

Data Index: 1536770 (Python Code)

Max Activating Token Index: 1012

Text #6

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.1560. Min Act: -0.1700

Data Index: 1360512 (Python Code)

Max Activating Token Index: 254

Text #7

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0935. Min Act: -0.1700

Data Index: 1583690 (Python Code)

Max Activating Token Index: 865

Text #8

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0623. Min Act: -0.1700

Data Index: 1617050 (Python Code)

Max Activating Token Index: 801

Text #9

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0623. Min Act: -0.1700

Data Index: 1645396 (Python Code)

Max Activating Token Index: 396

Text #10

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0623. Min Act: -0.1700

Data Index: 1645212 (Python Code)

Max Activating Token Index: 878

Text #11

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0623. Min Act: -0.1700

Data Index: 1383917 (Python Code)

Max Activating Token Index: 51

Text #12

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0935. Min Act: -0.1700

Data Index: 1505292 (Python Code)

Max Activating Token Index: 774

Text #13

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0310. Min Act: -0.1700

Data Index: 1503991 (Python Code)

Max Activating Token Index: 627

Text #14

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0310. Min Act: -0.1700

Data Index: 1408643 (Python Code)

Max Activating Token Index: 347

Text #15

Max Range: 5.2810. Min Range: -5.2810

Max Act: 4.9998. Min Act: -0.1700

Data Index: 1568465 (Python Code)

Max Activating Token Index: 564

Text #16

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0310. Min Act: -0.1700

Data Index: 1605589 (Python Code)

Max Activating Token Index: 48

Text #17

Max Range: 5.2810. Min Range: -5.2810

Max Act: 5.0623. Min Act: -0.1700

Data Index: 1385417 (Python Code)

Max Activating Token Index: 62

Text #18

Max Range: 5.2810. Min Range: -5.2810

Max Act: 4.9685. Min Act: -0.1700

Data Index: 1543060 (Python Code)

Max Activating Token Index: 61

Text #19

Max Range: 5.2810. Min Range: -5.2810

Max Act: 4.9998. Min Act: -0.1700

Data Index: 1400125 (Python Code)

Max Activating Token Index: 968