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

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

Neuron 1997 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: 3.8730. Min Range: -3.8730

Max Act: 3.8730. Min Act: -0.1700

Data Index: 867693 (C4 (Web Text))

Max Activating Token Index: 358

Text #1

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7792. Min Act: -0.1700

Data Index: 319607 (C4 (Web Text))

Max Activating Token Index: 266

Text #2

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7792. Min Act: -0.1700

Data Index: 1631961 (Python Code)

Max Activating Token Index: 759

Text #3

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7166. Min Act: -0.1700

Data Index: 568527 (C4 (Web Text))

Max Activating Token Index: 871

Text #4

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7166. Min Act: -0.1700

Data Index: 1183449 (C4 (Web Text))

Max Activating Token Index: 502

Text #5

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7166. Min Act: -0.1700

Data Index: 705155 (C4 (Web Text))

Max Activating Token Index: 134

Text #6

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7166. Min Act: -0.1700

Data Index: 415427 (C4 (Web Text))

Max Activating Token Index: 285

Text #7

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7479. Min Act: -0.1700

Data Index: 1056098 (C4 (Web Text))

Max Activating Token Index: 1014

Text #8

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7479. Min Act: -0.1700

Data Index: 1421175 (Python Code)

Max Activating Token Index: 203

Text #9

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.7166. Min Act: -0.1700

Data Index: 1155537 (C4 (Web Text))

Max Activating Token Index: 862

Text #10

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6853. Min Act: -0.1700

Data Index: 746175 (C4 (Web Text))

Max Activating Token Index: 60

Text #11

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6540. Min Act: -0.1700

Data Index: 43559 (C4 (Web Text))

Max Activating Token Index: 700

Text #12

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6853. Min Act: -0.1700

Data Index: 355840 (C4 (Web Text))

Max Activating Token Index: 290

Text #13

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6853. Min Act: -0.1700

Data Index: 245028 (C4 (Web Text))

Max Activating Token Index: 353

Text #14

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6853. Min Act: -0.1700

Data Index: 1353242 (C4 (Web Text))

Max Activating Token Index: 373

Text #15

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6540. Min Act: -0.1700

Data Index: 1192954 (C4 (Web Text))

Max Activating Token Index: 694

Text #16

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6853. Min Act: -0.1700

Data Index: 986319 (C4 (Web Text))

Max Activating Token Index: 463

Text #17

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6540. Min Act: -0.1700

Data Index: 1437467 (Python Code)

Max Activating Token Index: 1000

Text #18

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.5914. Min Act: -0.1700

Data Index: 1617254 (Python Code)

Max Activating Token Index: 707

Text #19

Max Range: 3.8730. Min Range: -3.8730

Max Act: 3.6227. Min Act: -0.1700

Data Index: 1626329 (Python Code)

Max Activating Token Index: 521