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

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

Neuron 1580 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: 7.5160. Min Range: -7.5160

Max Act: 7.5160. Min Act: -0.1689

Data Index: 1327531 (C4 (Web Text))

Max Activating Token Index: 186

Text #1

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.2660. Min Act: -0.1697

Data Index: 1075680 (C4 (Web Text))

Max Activating Token Index: 886

Text #2

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.3910. Min Act: -0.1700

Data Index: 834589 (C4 (Web Text))

Max Activating Token Index: 651

Text #3

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.3285. Min Act: -0.1700

Data Index: 990947 (C4 (Web Text))

Max Activating Token Index: 820

Text #4

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.2660. Min Act: -0.1697

Data Index: 1192020 (C4 (Web Text))

Max Activating Token Index: 868

Text #5

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.2660. Min Act: -0.1697

Data Index: 1320325 (C4 (Web Text))

Max Activating Token Index: 441

Text #6

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.2660. Min Act: -0.1700

Data Index: 422004 (C4 (Web Text))

Max Activating Token Index: 55

Text #7

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.2035. Min Act: -0.1691

Data Index: 501346 (C4 (Web Text))

Max Activating Token Index: 887

Text #8

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.2035. Min Act: -0.1700

Data Index: 250444 (C4 (Web Text))

Max Activating Token Index: 475

Text #9

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0785. Min Act: -0.1689

Data Index: 922363 (C4 (Web Text))

Max Activating Token Index: 827

Text #10

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.1410. Min Act: -0.1700

Data Index: 435839 (C4 (Web Text))

Max Activating Token Index: 838

Text #11

Max Range: 7.5160. Min Range: -7.5160

Max Act: 6.9848. Min Act: -0.1700

Data Index: 201976 (C4 (Web Text))

Max Activating Token Index: 78

Text #12

Max Range: 7.5160. Min Range: -7.5160

Max Act: 6.9848. Min Act: -0.1699

Data Index: 679812 (C4 (Web Text))

Max Activating Token Index: 529

Text #13

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0785. Min Act: -0.1700

Data Index: 888216 (C4 (Web Text))

Max Activating Token Index: 625

Text #14

Max Range: 7.5160. Min Range: -7.5160

Max Act: 6.9848. Min Act: -0.1689

Data Index: 539734 (C4 (Web Text))

Max Activating Token Index: 972

Text #15

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0785. Min Act: -0.1700

Data Index: 500872 (C4 (Web Text))

Max Activating Token Index: 650

Text #16

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0785. Min Act: -0.1700

Data Index: 390242 (C4 (Web Text))

Max Activating Token Index: 737

Text #17

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0160. Min Act: -0.1700

Data Index: 110064 (C4 (Web Text))

Max Activating Token Index: 634

Text #18

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0160. Min Act: -0.1700

Data Index: 1125322 (C4 (Web Text))

Max Activating Token Index: 394

Text #19

Max Range: 7.5160. Min Range: -7.5160

Max Act: 7.0160. Min Act: -0.1699

Data Index: 771173 (C4 (Web Text))

Max Activating Token Index: 276