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

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

Neuron 570 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.5064. Min Act: -0.1700

Data Index: 612161 (C4 (Web Text))

Max Activating Token Index: 943

Text #1

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3970. Min Act: -0.1700

Data Index: 906076 (C4 (Web Text))

Max Activating Token Index: 979

Text #2

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3814. Min Act: -0.1700

Data Index: 625767 (C4 (Web Text))

Max Activating Token Index: 93

Text #3

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3657. Min Act: -0.1700

Data Index: 446759 (C4 (Web Text))

Max Activating Token Index: 574

Text #4

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3032. Min Act: -0.1700

Data Index: 794775 (C4 (Web Text))

Max Activating Token Index: 676

Text #5

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2876. Min Act: -0.1700

Data Index: 321003 (C4 (Web Text))

Max Activating Token Index: 244

Text #6

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2564. Min Act: -0.1700

Data Index: 427584 (C4 (Web Text))

Max Activating Token Index: 11

Text #7

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3032. Min Act: -0.1700

Data Index: 331048 (C4 (Web Text))

Max Activating Token Index: 833

Text #8

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3189. Min Act: -0.1700

Data Index: 262616 (C4 (Web Text))

Max Activating Token Index: 283

Text #9

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3032. Min Act: -0.1700

Data Index: 1057761 (C4 (Web Text))

Max Activating Token Index: 790

Text #10

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2564. Min Act: -0.1700

Data Index: 1118475 (C4 (Web Text))

Max Activating Token Index: 972

Text #11

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2876. Min Act: -0.1700

Data Index: 494375 (C4 (Web Text))

Max Activating Token Index: 217

Text #12

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.3189. Min Act: -0.1700

Data Index: 514354 (C4 (Web Text))

Max Activating Token Index: 942

Text #13

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.1782. Min Act: -0.1700

Data Index: 1299639 (C4 (Web Text))

Max Activating Token Index: 954

Text #14

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2564. Min Act: -0.1700

Data Index: 357219 (C4 (Web Text))

Max Activating Token Index: 861

Text #15

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.1782. Min Act: -0.1700

Data Index: 215634 (C4 (Web Text))

Max Activating Token Index: 289

Text #16

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2095. Min Act: -0.1700

Data Index: 1143588 (C4 (Web Text))

Max Activating Token Index: 206

Text #17

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.1626. Min Act: -0.1700

Data Index: 841957 (C4 (Web Text))

Max Activating Token Index: 800

Text #18

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2564. Min Act: -0.1700

Data Index: 1053489 (C4 (Web Text))

Max Activating Token Index: 876

Text #19

Max Range: 5.5064. Min Range: -5.5064

Max Act: 5.2095. Min Act: -0.1700

Data Index: 188543 (C4 (Web Text))

Max Activating Token Index: 816