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

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

Neuron 2003 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: 5.4722. Min Range: -5.4722

Max Act: 5.4722. Min Act: -0.1700

Data Index: 481228 (C4 (Web Text))

Max Activating Token Index: 964

Text #1

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.1910. Min Act: -0.1700

Data Index: 792743 (C4 (Web Text))

Max Activating Token Index: 839

Text #2

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0972. Min Act: -0.1700

Data Index: 289669 (C4 (Web Text))

Max Activating Token Index: 638

Text #3

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0972. Min Act: -0.1700

Data Index: 130194 (C4 (Web Text))

Max Activating Token Index: 931

Text #4

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0972. Min Act: -0.1700

Data Index: 676078 (C4 (Web Text))

Max Activating Token Index: 29

Text #5

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0035. Min Act: -0.1700

Data Index: 256132 (C4 (Web Text))

Max Activating Token Index: 444

Text #6

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0347. Min Act: -0.1700

Data Index: 264904 (C4 (Web Text))

Max Activating Token Index: 269

Text #7

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.9722. Min Act: -0.1700

Data Index: 680639 (C4 (Web Text))

Max Activating Token Index: 891

Text #8

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0035. Min Act: -0.1700

Data Index: 782259 (C4 (Web Text))

Max Activating Token Index: 275

Text #9

Max Range: 5.4722. Min Range: -5.4722

Max Act: 5.0035. Min Act: -0.1700

Data Index: 1121206 (C4 (Web Text))

Max Activating Token Index: 1005

Text #10

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.9097. Min Act: -0.1700

Data Index: 345153 (C4 (Web Text))

Max Activating Token Index: 819

Text #11

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.9410. Min Act: -0.1700

Data Index: 311775 (C4 (Web Text))

Max Activating Token Index: 928

Text #12

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.9410. Min Act: -0.1700

Data Index: 515496 (C4 (Web Text))

Max Activating Token Index: 366

Text #13

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.9097. Min Act: -0.1700

Data Index: 592681 (C4 (Web Text))

Max Activating Token Index: 307

Text #14

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.8785. Min Act: -0.1700

Data Index: 738764 (C4 (Web Text))

Max Activating Token Index: 66

Text #15

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.8785. Min Act: -0.1700

Data Index: 1202747 (C4 (Web Text))

Max Activating Token Index: 1006

Text #16

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.9097. Min Act: -0.1700

Data Index: 1111910 (C4 (Web Text))

Max Activating Token Index: 195

Text #17

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.8472. Min Act: -0.1700

Data Index: 670295 (C4 (Web Text))

Max Activating Token Index: 925

Text #18

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.8785. Min Act: -0.1700

Data Index: 46964 (C4 (Web Text))

Max Activating Token Index: 246

Text #19

Max Range: 5.4722. Min Range: -5.4722

Max Act: 4.8472. Min Act: -0.1700

Data Index: 388791 (C4 (Web Text))

Max Activating Token Index: 955