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

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

Neuron 862 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: 7.8495. Min Range: -7.8495

Max Act: 7.8495. Min Act: -0.1700

Data Index: 1435802 (Python Code)

Max Activating Token Index: 332

Text #1

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.7245. Min Act: -0.1700

Data Index: 1595184 (Python Code)

Max Activating Token Index: 68

Text #2

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.5370. Min Act: -0.1700

Data Index: 1555057 (Python Code)

Max Activating Token Index: 1018

Text #3

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.4120. Min Act: -0.1700

Data Index: 833546 (C4 (Web Text))

Max Activating Token Index: 905

Text #4

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.4745. Min Act: -0.1700

Data Index: 1400340 (Python Code)

Max Activating Token Index: 835

Text #5

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.2870. Min Act: -0.1700

Data Index: 1420362 (Python Code)

Max Activating Token Index: 226

Text #6

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.1620. Min Act: -0.1700

Data Index: 1564845 (Python Code)

Max Activating Token Index: 846

Text #7

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.0370. Min Act: -0.1700

Data Index: 1620059 (Python Code)

Max Activating Token Index: 492

Text #8

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.1620. Min Act: -0.1700

Data Index: 1592469 (Python Code)

Max Activating Token Index: 333

Text #9

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.0370. Min Act: -0.1700

Data Index: 1656094 (Python Code)

Max Activating Token Index: 502

Text #10

Max Range: 7.8495. Min Range: -7.8495

Max Act: 7.0995. Min Act: -0.1700

Data Index: 1514597 (Python Code)

Max Activating Token Index: 739

Text #11

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.9745. Min Act: -0.1700

Data Index: 1605751 (Python Code)

Max Activating Token Index: 903

Text #12

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.9745. Min Act: -0.1700

Data Index: 1595062 (Python Code)

Max Activating Token Index: 740

Text #13

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.9120. Min Act: -0.1700

Data Index: 1367199 (Python Code)

Max Activating Token Index: 751

Text #14

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.8495. Min Act: -0.1700

Data Index: 1566683 (Python Code)

Max Activating Token Index: 924

Text #15

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.7245. Min Act: -0.1700

Data Index: 1378207 (Python Code)

Max Activating Token Index: 497

Text #16

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.6620. Min Act: -0.1700

Data Index: 1471416 (Python Code)

Max Activating Token Index: 108

Text #17

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.9120. Min Act: -0.1700

Data Index: 1394210 (Python Code)

Max Activating Token Index: 90

Text #18

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.6620. Min Act: -0.1700

Data Index: 1527717 (Python Code)

Max Activating Token Index: 712

Text #19

Max Range: 7.8495. Min Range: -7.8495

Max Act: 6.6620. Min Act: -0.1700

Data Index: 1524332 (Python Code)

Max Activating Token Index: 166