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

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

Neuron 1924 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: 3.8472. Min Range: -3.8472

Max Act: 3.8472. Min Act: -0.1700

Data Index: 364274 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.8316. Min Act: -0.1700

Data Index: 1224094 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.8003. Min Act: -0.1700

Data Index: 854093 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.6751. Min Act: -0.1700

Data Index: 300565 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.6438. Min Act: -0.1700

Data Index: 988010 (C4 (Web Text))

Max Activating Token Index: 5

Text #5

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.6125. Min Act: -0.1700

Data Index: 1467032 (Python Code)

Max Activating Token Index: 604

Text #6

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5969. Min Act: -0.1700

Data Index: 1206120 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.6125. Min Act: -0.1700

Data Index: 621763 (C4 (Web Text))

Max Activating Token Index: 11

Text #8

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5969. Min Act: -0.1700

Data Index: 1348058 (C4 (Web Text))

Max Activating Token Index: 61

Text #9

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5812. Min Act: -0.1700

Data Index: 327178 (C4 (Web Text))

Max Activating Token Index: 22

Text #10

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5655. Min Act: -0.1700

Data Index: 1653455 (Python Code)

Max Activating Token Index: 405

Text #11

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5812. Min Act: -0.1700

Data Index: 1640742 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5655. Min Act: -0.1700

Data Index: 1532735 (Python Code)

Max Activating Token Index: 41

Text #13

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5655. Min Act: -0.1700

Data Index: 1198451 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5655. Min Act: -0.1700

Data Index: 1468673 (Python Code)

Max Activating Token Index: 565

Text #15

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5342. Min Act: -0.1700

Data Index: 1626196 (Python Code)

Max Activating Token Index: 159

Text #16

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5499. Min Act: -0.1700

Data Index: 880748 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5342. Min Act: -0.1700

Data Index: 1649277 (Python Code)

Max Activating Token Index: 130

Text #18

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5342. Min Act: -0.1700

Data Index: 1401727 (Python Code)

Max Activating Token Index: 471

Text #19

Max Range: 3.8472. Min Range: -3.8472

Max Act: 3.5342. Min Act: -0.1700

Data Index: 1635184 (Python Code)

Max Activating Token Index: 259