Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1694 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.0375. Min Range: -3.0375

Max Act: 3.0375. Min Act: -0.1700

Data Index: 1426120 (Python Code)

Max Activating Token Index: 897

Text #1

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9743. Min Act: -0.1700

Data Index: 713790 (C4 (Web Text))

Max Activating Token Index: 767

Text #2

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9743. Min Act: -0.1700

Data Index: 1146088 (C4 (Web Text))

Max Activating Token Index: 921

Text #3

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9426. Min Act: -0.1700

Data Index: 735289 (C4 (Web Text))

Max Activating Token Index: 334

Text #4

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9584. Min Act: -0.1700

Data Index: 115704 (C4 (Web Text))

Max Activating Token Index: 100

Text #5

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9268. Min Act: -0.1700

Data Index: 220765 (C4 (Web Text))

Max Activating Token Index: 82

Text #6

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9109. Min Act: -0.1700

Data Index: 853063 (C4 (Web Text))

Max Activating Token Index: 78

Text #7

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9268. Min Act: -0.1700

Data Index: 650533 (C4 (Web Text))

Max Activating Token Index: 672

Text #8

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.9268. Min Act: -0.1700

Data Index: 708870 (C4 (Web Text))

Max Activating Token Index: 148

Text #9

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8950. Min Act: -0.1700

Data Index: 64485 (C4 (Web Text))

Max Activating Token Index: 718

Text #10

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8633. Min Act: -0.1700

Data Index: 11062 (C4 (Web Text))

Max Activating Token Index: 303

Text #11

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8633. Min Act: -0.1700

Data Index: 722138 (C4 (Web Text))

Max Activating Token Index: 917

Text #12

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8792. Min Act: -0.1700

Data Index: 742242 (C4 (Web Text))

Max Activating Token Index: 639

Text #13

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8792. Min Act: -0.1700

Data Index: 236533 (C4 (Web Text))

Max Activating Token Index: 669

Text #14

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8633. Min Act: -0.1700

Data Index: 731926 (C4 (Web Text))

Max Activating Token Index: 113

Text #15

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8792. Min Act: -0.1700

Data Index: 1343557 (C4 (Web Text))

Max Activating Token Index: 34

Text #16

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8633. Min Act: -0.1700

Data Index: 423985 (C4 (Web Text))

Max Activating Token Index: 850

Text #17

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8474. Min Act: -0.1700

Data Index: 1587557 (Python Code)

Max Activating Token Index: 142

Text #18

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8474. Min Act: -0.1700

Data Index: 1122647 (C4 (Web Text))

Max Activating Token Index: 99

Text #19

Max Range: 3.0375. Min Range: -3.0375

Max Act: 2.8633. Min Act: -0.1700

Data Index: 433793 (C4 (Web Text))

Max Activating Token Index: 259