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

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

Neuron 587 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.7095. Min Act: -0.1700

Data Index: 1648636 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.6939. Min Act: -0.1700

Data Index: 1487869 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.6469. Min Act: -0.1700

Data Index: 448961 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.5687. Min Act: -0.1700

Data Index: 1559746 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.5687. Min Act: -0.1700

Data Index: 1456013 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.6156. Min Act: -0.1700

Data Index: 1620720 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.4903. Min Act: -0.1700

Data Index: 1522717 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.4903. Min Act: -0.1700

Data Index: 1388241 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.4590. Min Act: -0.1700

Data Index: 1539189 (Python Code)

Max Activating Token Index: 1023

Text #9

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.4276. Min Act: -0.1700

Data Index: 1181803 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.4747. Min Act: -0.1700

Data Index: 1381103 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.3649. Min Act: -0.1700

Data Index: 1392781 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.3178. Min Act: -0.1700

Data Index: 1397870 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.2707. Min Act: -0.1700

Data Index: 1602855 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.2707. Min Act: -0.1700

Data Index: 1476510 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.2707. Min Act: -0.1700

Data Index: 1647870 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.2392. Min Act: -0.1700

Data Index: 399238 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.2392. Min Act: -0.1700

Data Index: 1594131 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.1921. Min Act: -0.1700

Data Index: 1501808 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 3.7095. Min Range: -3.7095

Max Act: 3.1921. Min Act: -0.1700

Data Index: 1489910 (Python Code)

Max Activating Token Index: 1023