Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 1938 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.9984. Min Act: -0.1700

Data Index: 345978 (C4 (Web Text))

Max Activating Token Index: 967

Text #1

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.5453. Min Act: -0.1700

Data Index: 173602 (C4 (Web Text))

Max Activating Token Index: 867

Text #2

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.5297. Min Act: -0.1700

Data Index: 112534 (C4 (Web Text))

Max Activating Token Index: 812

Text #3

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.4828. Min Act: -0.1700

Data Index: 769946 (C4 (Web Text))

Max Activating Token Index: 700

Text #4

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.4515. Min Act: -0.1700

Data Index: 223605 (C4 (Web Text))

Max Activating Token Index: 580

Text #5

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.3265. Min Act: -0.1700

Data Index: 709140 (C4 (Web Text))

Max Activating Token Index: 941

Text #6

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.3109. Min Act: -0.1700

Data Index: 326498 (C4 (Web Text))

Max Activating Token Index: 302

Text #7

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.2640. Min Act: -0.1700

Data Index: 467973 (C4 (Web Text))

Max Activating Token Index: 584

Text #8

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.1859. Min Act: -0.1700

Data Index: 1201275 (C4 (Web Text))

Max Activating Token Index: 767

Text #9

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.1390. Min Act: -0.1700

Data Index: 1134256 (C4 (Web Text))

Max Activating Token Index: 332

Text #10

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.1234. Min Act: -0.1700

Data Index: 1226313 (C4 (Web Text))

Max Activating Token Index: 624

Text #11

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.8734. Min Act: -0.1700

Data Index: 1324353 (C4 (Web Text))

Max Activating Token Index: 766

Text #12

Max Range: 8.9984. Min Range: -8.9984

Max Act: 8.0297. Min Act: -0.1700

Data Index: 649746 (C4 (Web Text))

Max Activating Token Index: 888

Text #13

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.9828. Min Act: -0.1700

Data Index: 1353823 (C4 (Web Text))

Max Activating Token Index: 774

Text #14

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.9203. Min Act: -0.1700

Data Index: 234702 (C4 (Web Text))

Max Activating Token Index: 531

Text #15

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.8734. Min Act: -0.1700

Data Index: 159570 (C4 (Web Text))

Max Activating Token Index: 729

Text #16

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.9828. Min Act: -0.1700

Data Index: 975124 (C4 (Web Text))

Max Activating Token Index: 769

Text #17

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.7640. Min Act: -0.1700

Data Index: 682716 (C4 (Web Text))

Max Activating Token Index: 950

Text #18

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.8422. Min Act: -0.1700

Data Index: 917236 (C4 (Web Text))

Max Activating Token Index: 594

Text #19

Max Range: 8.9984. Min Range: -8.9984

Max Act: 7.7640. Min Act: -0.1700

Data Index: 1475416 (Python Code)

Max Activating Token Index: 970