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

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

Neuron 1191 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.3567. Min Act: -0.1700

Data Index: 562329 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.3254. Min Act: -0.1700

Data Index: 37272 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.1692. Min Act: -0.1700

Data Index: 1163090 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.2317. Min Act: -0.1700

Data Index: 414219 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.1067. Min Act: -0.1700

Data Index: 664850 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.1067. Min Act: -0.1700

Data Index: 1060241 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.0442. Min Act: -0.1700

Data Index: 764835 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.1379. Min Act: -0.1700

Data Index: 90390 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.1067. Min Act: -0.1700

Data Index: 1018259 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.0754. Min Act: -0.1700

Data Index: 591977 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 6.3567. Min Range: -6.3567

Max Act: 5.9817. Min Act: -0.1700

Data Index: 17630 (C4 (Web Text))

Max Activating Token Index: 257

Text #11

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.0754. Min Act: -0.1700

Data Index: 6126 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.1067. Min Act: -0.1700

Data Index: 165573 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.0442. Min Act: -0.1700

Data Index: 1014894 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.0442. Min Act: -0.1700

Data Index: 300405 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 6.3567. Min Range: -6.3567

Max Act: 5.9504. Min Act: -0.1700

Data Index: 235114 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 6.3567. Min Range: -6.3567

Max Act: 5.9817. Min Act: -0.1700

Data Index: 412221 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 6.3567. Min Range: -6.3567

Max Act: 6.0129. Min Act: -0.1700

Data Index: 1354816 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 6.3567. Min Range: -6.3567

Max Act: 5.9504. Min Act: -0.1700

Data Index: 183995 (C4 (Web Text))

Max Activating Token Index: 654

Text #19

Max Range: 6.3567. Min Range: -6.3567

Max Act: 5.8254. Min Act: -0.1700

Data Index: 593787 (C4 (Web Text))

Max Activating Token Index: 1023