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

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

Neuron 119 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: 3.2516. Min Range: -3.2516

Max Act: 3.2516. Min Act: -0.1700

Data Index: 905263 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.2359. Min Act: -0.1700

Data Index: 167342 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.1730. Min Act: -0.1700

Data Index: 277460 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.1100. Min Act: -0.1700

Data Index: 593422 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.1100. Min Act: -0.1700

Data Index: 265332 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.1257. Min Act: -0.1700

Data Index: 273927 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0942. Min Act: -0.1700

Data Index: 1568896 (Python Code)

Max Activating Token Index: 537

Text #7

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0469. Min Act: -0.1700

Data Index: 823057 (C4 (Web Text))

Max Activating Token Index: 638

Text #8

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0627. Min Act: -0.1700

Data Index: 1054323 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0311. Min Act: -0.1700

Data Index: 1588593 (Python Code)

Max Activating Token Index: 790

Text #10

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0469. Min Act: -0.1700

Data Index: 1447026 (Python Code)

Max Activating Token Index: 861

Text #11

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0469. Min Act: -0.1700

Data Index: 150897 (C4 (Web Text))

Max Activating Token Index: 18

Text #12

Max Range: 3.2516. Min Range: -3.2516

Max Act: 3.0311. Min Act: -0.1700

Data Index: 1457328 (Python Code)

Max Activating Token Index: 936

Text #13

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9995. Min Act: -0.1700

Data Index: 1466782 (Python Code)

Max Activating Token Index: 193

Text #14

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9678. Min Act: -0.1700

Data Index: 1137899 (C4 (Web Text))

Max Activating Token Index: 835

Text #15

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9362. Min Act: -0.1700

Data Index: 1558059 (Python Code)

Max Activating Token Index: 448

Text #16

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9995. Min Act: -0.1700

Data Index: 847977 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9837. Min Act: -0.1700

Data Index: 1430978 (Python Code)

Max Activating Token Index: 914

Text #18

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9520. Min Act: -0.1700

Data Index: 478299 (C4 (Web Text))

Max Activating Token Index: 873

Text #19

Max Range: 3.2516. Min Range: -3.2516

Max Act: 2.9362. Min Act: -0.1700

Data Index: 59743 (C4 (Web Text))

Max Activating Token Index: 480