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

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

Neuron 1505 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.3174. Min Act: -0.1700

Data Index: 216939 (C4 (Web Text))

Max Activating Token Index: 895

Text #1

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.1611. Min Act: -0.1700

Data Index: 754782 (C4 (Web Text))

Max Activating Token Index: 752

Text #2

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.1924. Min Act: -0.1700

Data Index: 95962 (C4 (Web Text))

Max Activating Token Index: 392

Text #3

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.0986. Min Act: -0.1700

Data Index: 1327257 (C4 (Web Text))

Max Activating Token Index: 735

Text #4

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.0674. Min Act: -0.1700

Data Index: 381491 (C4 (Web Text))

Max Activating Token Index: 868

Text #5

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.0361. Min Act: -0.1700

Data Index: 100904 (C4 (Web Text))

Max Activating Token Index: 819

Text #6

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.0361. Min Act: -0.1700

Data Index: 591209 (C4 (Web Text))

Max Activating Token Index: 338

Text #7

Max Range: 5.3174. Min Range: -5.3174

Max Act: 5.0049. Min Act: -0.1700

Data Index: 210608 (C4 (Web Text))

Max Activating Token Index: 317

Text #8

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.9736. Min Act: -0.1700

Data Index: 232483 (C4 (Web Text))

Max Activating Token Index: 35

Text #9

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.8799. Min Act: -0.1700

Data Index: 1606345 (Python Code)

Max Activating Token Index: 968

Text #10

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.9111. Min Act: -0.1700

Data Index: 26901 (C4 (Web Text))

Max Activating Token Index: 949

Text #11

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.9111. Min Act: -0.1700

Data Index: 732866 (C4 (Web Text))

Max Activating Token Index: 892

Text #12

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.9111. Min Act: -0.1700

Data Index: 58807 (C4 (Web Text))

Max Activating Token Index: 888

Text #13

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.9424. Min Act: -0.1700

Data Index: 431027 (C4 (Web Text))

Max Activating Token Index: 753

Text #14

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.9111. Min Act: -0.1700

Data Index: 151512 (C4 (Web Text))

Max Activating Token Index: 381

Text #15

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.8799. Min Act: -0.1700

Data Index: 674108 (C4 (Web Text))

Max Activating Token Index: 45

Text #16

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.8799. Min Act: -0.1700

Data Index: 1096400 (C4 (Web Text))

Max Activating Token Index: 539

Text #17

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.8799. Min Act: -0.1700

Data Index: 193370 (C4 (Web Text))

Max Activating Token Index: 347

Text #18

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.8799. Min Act: -0.1700

Data Index: 516000 (C4 (Web Text))

Max Activating Token Index: 127

Text #19

Max Range: 5.3174. Min Range: -5.3174

Max Act: 4.8486. Min Act: -0.1700

Data Index: 72349 (C4 (Web Text))

Max Activating Token Index: 916