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

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

Neuron 1739 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.1767. Min Range: -3.1767

Max Act: 3.1767. Min Act: -0.1700

Data Index: 855360 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.1767. Min Range: -3.1767

Max Act: 3.0506. Min Act: -0.1700

Data Index: 1217181 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.1767. Min Range: -3.1767

Max Act: 3.0506. Min Act: -0.1700

Data Index: 322186 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.9399. Min Act: -0.1700

Data Index: 377618 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.9557. Min Act: -0.1700

Data Index: 1163752 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.9240. Min Act: -0.1700

Data Index: 644659 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8606. Min Act: -0.1700

Data Index: 1144812 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8447. Min Act: -0.1700

Data Index: 1010442 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8288. Min Act: -0.1700

Data Index: 1034513 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8129. Min Act: -0.1700

Data Index: 1041756 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8129. Min Act: -0.1700

Data Index: 965546 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8288. Min Act: -0.1700

Data Index: 897951 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.8129. Min Act: -0.1700

Data Index: 1399962 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7969. Min Act: -0.1700

Data Index: 412357 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7810. Min Act: -0.1700

Data Index: 58382 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7969. Min Act: -0.1700

Data Index: 1087088 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7810. Min Act: -0.1700

Data Index: 1636812 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7810. Min Act: -0.1700

Data Index: 33930 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7491. Min Act: -0.1700

Data Index: 1423734 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 3.1767. Min Range: -3.1767

Max Act: 2.7491. Min Act: -0.1700

Data Index: 274592 (C4 (Web Text))

Max Activating Token Index: 1023