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

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

Neuron 136 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: 5.2673. Min Range: -5.2673

Max Act: 5.2673. Min Act: -0.1700

Data Index: 1622422 (Python Code)

Max Activating Token Index: 225

Text #1

Max Range: 5.2673. Min Range: -5.2673

Max Act: 5.2048. Min Act: -0.1700

Data Index: 1590921 (Python Code)

Max Activating Token Index: 195

Text #2

Max Range: 5.2673. Min Range: -5.2673

Max Act: 5.1423. Min Act: -0.1700

Data Index: 1388826 (Python Code)

Max Activating Token Index: 247

Text #3

Max Range: 5.2673. Min Range: -5.2673

Max Act: 5.1111. Min Act: -0.1700

Data Index: 1585151 (Python Code)

Max Activating Token Index: 910

Text #4

Max Range: 5.2673. Min Range: -5.2673

Max Act: 5.1423. Min Act: -0.1700

Data Index: 1654389 (Python Code)

Max Activating Token Index: 1009

Text #5

Max Range: 5.2673. Min Range: -5.2673

Max Act: 5.1111. Min Act: -0.1700

Data Index: 1549480 (Python Code)

Max Activating Token Index: 498

Text #6

Max Range: 5.2673. Min Range: -5.2673

Max Act: 5.0173. Min Act: -0.1700

Data Index: 1605647 (Python Code)

Max Activating Token Index: 918

Text #7

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.9861. Min Act: -0.1700

Data Index: 1579039 (Python Code)

Max Activating Token Index: 59

Text #8

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.8611. Min Act: -0.1700

Data Index: 1541948 (Python Code)

Max Activating Token Index: 826

Text #9

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.8923. Min Act: -0.1700

Data Index: 1426896 (Python Code)

Max Activating Token Index: 121

Text #10

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.8923. Min Act: -0.1700

Data Index: 1538209 (Python Code)

Max Activating Token Index: 837

Text #11

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.8611. Min Act: -0.1700

Data Index: 1543832 (Python Code)

Max Activating Token Index: 991

Text #12

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7673. Min Act: -0.1700

Data Index: 1550152 (Python Code)

Max Activating Token Index: 783

Text #13

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.8298. Min Act: -0.1700

Data Index: 1368983 (Python Code)

Max Activating Token Index: 147

Text #14

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7673. Min Act: -0.1700

Data Index: 1639491 (Python Code)

Max Activating Token Index: 699

Text #15

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7048. Min Act: -0.1700

Data Index: 1427439 (Python Code)

Max Activating Token Index: 975

Text #16

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7361. Min Act: -0.1700

Data Index: 1614612 (Python Code)

Max Activating Token Index: 1020

Text #17

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7673. Min Act: -0.1700

Data Index: 1473139 (Python Code)

Max Activating Token Index: 470

Text #18

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7673. Min Act: -0.1700

Data Index: 1438897 (Python Code)

Max Activating Token Index: 916

Text #19

Max Range: 5.2673. Min Range: -5.2673

Max Act: 4.7673. Min Act: -0.1700

Data Index: 1543925 (Python Code)

Max Activating Token Index: 134