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

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

Neuron 1645 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: 3.8960. Min Range: -3.8960

Max Act: 3.8960. Min Act: -0.1700

Data Index: 1301508 (C4 (Web Text))

Max Activating Token Index: 610

Text #1

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8803. Min Act: -0.1700

Data Index: 1293921 (C4 (Web Text))

Max Activating Token Index: 508

Text #2

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8647. Min Act: -0.1700

Data Index: 1282176 (C4 (Web Text))

Max Activating Token Index: 528

Text #3

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8490. Min Act: -0.1700

Data Index: 696344 (C4 (Web Text))

Max Activating Token Index: 191

Text #4

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8178. Min Act: -0.1700

Data Index: 909845 (C4 (Web Text))

Max Activating Token Index: 516

Text #5

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8647. Min Act: -0.1700

Data Index: 1260763 (C4 (Web Text))

Max Activating Token Index: 629

Text #6

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8334. Min Act: -0.1700

Data Index: 1310448 (C4 (Web Text))

Max Activating Token Index: 799

Text #7

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8021. Min Act: -0.1700

Data Index: 1501918 (Python Code)

Max Activating Token Index: 869

Text #8

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7708. Min Act: -0.1700

Data Index: 1388458 (Python Code)

Max Activating Token Index: 185

Text #9

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8178. Min Act: -0.1700

Data Index: 704020 (C4 (Web Text))

Max Activating Token Index: 455

Text #10

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7552. Min Act: -0.1700

Data Index: 1238841 (C4 (Web Text))

Max Activating Token Index: 432

Text #11

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7865. Min Act: -0.1700

Data Index: 1270118 (C4 (Web Text))

Max Activating Token Index: 1004

Text #12

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.8178. Min Act: -0.1700

Data Index: 856808 (C4 (Web Text))

Max Activating Token Index: 951

Text #13

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7708. Min Act: -0.1700

Data Index: 716796 (C4 (Web Text))

Max Activating Token Index: 600

Text #14

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7865. Min Act: -0.1700

Data Index: 25130 (C4 (Web Text))

Max Activating Token Index: 272

Text #15

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7239. Min Act: -0.1700

Data Index: 609901 (C4 (Web Text))

Max Activating Token Index: 449

Text #16

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7552. Min Act: -0.1700

Data Index: 458723 (C4 (Web Text))

Max Activating Token Index: 38

Text #17

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7239. Min Act: -0.1700

Data Index: 903010 (C4 (Web Text))

Max Activating Token Index: 114

Text #18

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.7083. Min Act: -0.1700

Data Index: 524154 (C4 (Web Text))

Max Activating Token Index: 414

Text #19

Max Range: 3.8960. Min Range: -3.8960

Max Act: 3.6926. Min Act: -0.1700

Data Index: 307268 (C4 (Web Text))

Max Activating Token Index: 963