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

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

Neuron 535 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: 3.4070. Min Range: -3.4070

Max Act: 3.4070. Min Act: -0.1700

Data Index: 579678 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.2500. Min Act: -0.1700

Data Index: 1346717 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.1871. Min Act: -0.1700

Data Index: 1531295 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.1714. Min Act: -0.1700

Data Index: 1216712 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.1241. Min Act: -0.1700

Data Index: 1361684 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.1084. Min Act: -0.1700

Data Index: 331592 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.0768. Min Act: -0.1700

Data Index: 1295647 (C4 (Web Text))

Max Activating Token Index: 579

Text #7

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.0611. Min Act: -0.1700

Data Index: 403137 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.0768. Min Act: -0.1700

Data Index: 864445 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.0611. Min Act: -0.1700

Data Index: 10213 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.0295. Min Act: -0.1700

Data Index: 1487226 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 3.4070. Min Range: -3.4070

Max Act: 3.0295. Min Act: -0.1700

Data Index: 1631986 (Python Code)

Max Activating Token Index: 565

Text #12

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9979. Min Act: -0.1700

Data Index: 981870 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9662. Min Act: -0.1700

Data Index: 390312 (C4 (Web Text))

Max Activating Token Index: 1015

Text #14

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9504. Min Act: -0.1700

Data Index: 883418 (C4 (Web Text))

Max Activating Token Index: 987

Text #15

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9662. Min Act: -0.1700

Data Index: 881340 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9821. Min Act: -0.1700

Data Index: 1221981 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9346. Min Act: -0.1700

Data Index: 1216761 (C4 (Web Text))

Max Activating Token Index: 938

Text #18

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9504. Min Act: -0.1700

Data Index: 721889 (C4 (Web Text))

Max Activating Token Index: 977

Text #19

Max Range: 3.4070. Min Range: -3.4070

Max Act: 2.9346. Min Act: -0.1700

Data Index: 851836 (C4 (Web Text))

Max Activating Token Index: 1023