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

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

Neuron 1699 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: 3.5721. Min Range: -3.5721

Max Act: 3.5721. Min Act: -0.1700

Data Index: 1544436 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.4938. Min Act: -0.1700

Data Index: 347413 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.3056. Min Act: -0.1700

Data Index: 1449854 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.3840. Min Act: -0.1700

Data Index: 140308 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2742. Min Act: -0.1700

Data Index: 1587822 (Python Code)

Max Activating Token Index: 1023

Text #5

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2899. Min Act: -0.1700

Data Index: 1431147 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2742. Min Act: -0.1700

Data Index: 1362020 (Python Code)

Max Activating Token Index: 1023

Text #7

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.3056. Min Act: -0.1700

Data Index: 1330298 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.3056. Min Act: -0.1700

Data Index: 969759 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2742. Min Act: -0.1700

Data Index: 1442586 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2742. Min Act: -0.1700

Data Index: 1447695 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.3056. Min Act: -0.1700

Data Index: 754517 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2742. Min Act: -0.1700

Data Index: 1324990 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2899. Min Act: -0.1700

Data Index: 836438 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2584. Min Act: -0.1700

Data Index: 1534989 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2427. Min Act: -0.1700

Data Index: 1220934 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2113. Min Act: -0.1700

Data Index: 1513790 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2113. Min Act: -0.1700

Data Index: 1501866 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2113. Min Act: -0.1700

Data Index: 81415 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 3.5721. Min Range: -3.5721

Max Act: 3.2742. Min Act: -0.1700

Data Index: 342948 (C4 (Web Text))

Max Activating Token Index: 1023