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

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

Neuron 644 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.4235. Min Act: -0.1700

Data Index: 167455 (C4 (Web Text))

Max Activating Token Index: 97

Text #1

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2665. Min Act: -0.1700

Data Index: 663721 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.3293. Min Act: -0.1700

Data Index: 558464 (C4 (Web Text))

Max Activating Token Index: 857

Text #3

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2351. Min Act: -0.1700

Data Index: 1113708 (C4 (Web Text))

Max Activating Token Index: 648

Text #4

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2665. Min Act: -0.1700

Data Index: 34094 (C4 (Web Text))

Max Activating Token Index: 951

Text #5

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2665. Min Act: -0.1700

Data Index: 1159029 (C4 (Web Text))

Max Activating Token Index: 157

Text #6

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1721. Min Act: -0.1700

Data Index: 375310 (C4 (Web Text))

Max Activating Token Index: 10

Text #7

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2351. Min Act: -0.1700

Data Index: 879192 (C4 (Web Text))

Max Activating Token Index: 755

Text #8

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2036. Min Act: -0.1700

Data Index: 594258 (C4 (Web Text))

Max Activating Token Index: 373

Text #9

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2036. Min Act: -0.1700

Data Index: 739194 (C4 (Web Text))

Max Activating Token Index: 44

Text #10

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1721. Min Act: -0.1700

Data Index: 1062484 (C4 (Web Text))

Max Activating Token Index: 854

Text #11

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1406. Min Act: -0.1700

Data Index: 114657 (C4 (Web Text))

Max Activating Token Index: 707

Text #12

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1406. Min Act: -0.1700

Data Index: 335875 (C4 (Web Text))

Max Activating Token Index: 233

Text #13

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1406. Min Act: -0.1700

Data Index: 147847 (C4 (Web Text))

Max Activating Token Index: 415

Text #14

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2036. Min Act: -0.1700

Data Index: 334256 (C4 (Web Text))

Max Activating Token Index: 19

Text #15

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1721. Min Act: -0.1700

Data Index: 699792 (C4 (Web Text))

Max Activating Token Index: 879

Text #16

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1406. Min Act: -0.1700

Data Index: 101962 (C4 (Web Text))

Max Activating Token Index: 567

Text #17

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1721. Min Act: -0.1700

Data Index: 804613 (C4 (Web Text))

Max Activating Token Index: 695

Text #18

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.1406. Min Act: -0.1700

Data Index: 484901 (C4 (Web Text))

Max Activating Token Index: 669

Text #19

Max Range: 3.4235. Min Range: -3.4235

Max Act: 3.2036. Min Act: -0.1700

Data Index: 186185 (C4 (Web Text))

Max Activating Token Index: 10