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

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

Neuron 190 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.1389. Min Act: -0.1700

Data Index: 1526941 (Python Code)

Max Activating Token Index: 4

Text #1

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0916. Min Act: -0.1700

Data Index: 997631 (C4 (Web Text))

Max Activating Token Index: 451

Text #2

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0758. Min Act: -0.1700

Data Index: 1043208 (C4 (Web Text))

Max Activating Token Index: 836

Text #3

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0600. Min Act: -0.1700

Data Index: 1449537 (Python Code)

Max Activating Token Index: 29

Text #4

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0443. Min Act: -0.1700

Data Index: 1610148 (Python Code)

Max Activating Token Index: 203

Text #5

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0443. Min Act: -0.1700

Data Index: 1546366 (Python Code)

Max Activating Token Index: 93

Text #6

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0443. Min Act: -0.1700

Data Index: 1528815 (Python Code)

Max Activating Token Index: 22

Text #7

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0443. Min Act: -0.1700

Data Index: 1553489 (Python Code)

Max Activating Token Index: 15

Text #8

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0285. Min Act: -0.1700

Data Index: 552070 (C4 (Web Text))

Max Activating Token Index: 222

Text #9

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0285. Min Act: -0.1700

Data Index: 1415249 (Python Code)

Max Activating Token Index: 29

Text #10

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0443. Min Act: -0.1700

Data Index: 1613741 (Python Code)

Max Activating Token Index: 894

Text #11

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0285. Min Act: -0.1700

Data Index: 1464377 (Python Code)

Max Activating Token Index: 893

Text #12

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0127. Min Act: -0.1700

Data Index: 1615439 (Python Code)

Max Activating Token Index: 873

Text #13

Max Range: 3.1389. Min Range: -3.1389

Max Act: 2.9969. Min Act: -0.1700

Data Index: 1384624 (Python Code)

Max Activating Token Index: 153

Text #14

Max Range: 3.1389. Min Range: -3.1389

Max Act: 2.9969. Min Act: -0.1700

Data Index: 1632564 (Python Code)

Max Activating Token Index: 90

Text #15

Max Range: 3.1389. Min Range: -3.1389

Max Act: 3.0127. Min Act: -0.1700

Data Index: 405777 (C4 (Web Text))

Max Activating Token Index: 39

Text #16

Max Range: 3.1389. Min Range: -3.1389

Max Act: 2.9969. Min Act: -0.1700

Data Index: 1398489 (Python Code)

Max Activating Token Index: 781

Text #17

Max Range: 3.1389. Min Range: -3.1389

Max Act: 2.9810. Min Act: -0.1700

Data Index: 1476834 (Python Code)

Max Activating Token Index: 95

Text #18

Max Range: 3.1389. Min Range: -3.1389

Max Act: 2.9810. Min Act: -0.1700

Data Index: 982379 (C4 (Web Text))

Max Activating Token Index: 412

Text #19

Max Range: 3.1389. Min Range: -3.1389

Max Act: 2.9494. Min Act: -0.1700

Data Index: 473199 (C4 (Web Text))

Max Activating Token Index: 152