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

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

Neuron 1832 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: 6.5516. Min Range: -6.5516

Max Act: 6.5516. Min Act: -0.1700

Data Index: 378713 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.4266. Min Act: -0.1700

Data Index: 381441 (C4 (Web Text))

Max Activating Token Index: 981

Text #2

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.3016. Min Act: -0.1700

Data Index: 56137 (C4 (Web Text))

Max Activating Token Index: 880

Text #3

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.2079. Min Act: -0.1700

Data Index: 600355 (C4 (Web Text))

Max Activating Token Index: 981

Text #4

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1766. Min Act: -0.1700

Data Index: 328973 (C4 (Web Text))

Max Activating Token Index: 647

Text #5

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1766. Min Act: -0.1700

Data Index: 155010 (C4 (Web Text))

Max Activating Token Index: 627

Text #6

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1454. Min Act: -0.1700

Data Index: 1395242 (Python Code)

Max Activating Token Index: 878

Text #7

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1766. Min Act: -0.1700

Data Index: 1382500 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1454. Min Act: -0.1700

Data Index: 1000097 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1766. Min Act: -0.1700

Data Index: 685332 (C4 (Web Text))

Max Activating Token Index: 553

Text #10

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1141. Min Act: -0.1700

Data Index: 831942 (C4 (Web Text))

Max Activating Token Index: 587

Text #11

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1141. Min Act: -0.1700

Data Index: 1188550 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1141. Min Act: -0.1700

Data Index: 1418746 (Python Code)

Max Activating Token Index: 773

Text #13

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.0829. Min Act: -0.1700

Data Index: 275053 (C4 (Web Text))

Max Activating Token Index: 458

Text #14

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.1141. Min Act: -0.1700

Data Index: 100900 (C4 (Web Text))

Max Activating Token Index: 979

Text #15

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.0516. Min Act: -0.1700

Data Index: 580483 (C4 (Web Text))

Max Activating Token Index: 760

Text #16

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.0516. Min Act: -0.1699

Data Index: 1519236 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 6.5516. Min Range: -6.5516

Max Act: 5.9891. Min Act: -0.1700

Data Index: 1637884 (Python Code)

Max Activating Token Index: 778

Text #18

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.0204. Min Act: -0.1700

Data Index: 764171 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 6.5516. Min Range: -6.5516

Max Act: 6.0516. Min Act: -0.1700

Data Index: 121602 (C4 (Web Text))

Max Activating Token Index: 981