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

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

Neuron 1590 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.2864. Min Range: -5.2864

Max Act: 5.2864. Min Act: -0.1700

Data Index: 345628 (C4 (Web Text))

Max Activating Token Index: 616

Text #1

Max Range: 5.2864. Min Range: -5.2864

Max Act: 5.1926. Min Act: -0.1700

Data Index: 1128053 (C4 (Web Text))

Max Activating Token Index: 719

Text #2

Max Range: 5.2864. Min Range: -5.2864

Max Act: 5.1301. Min Act: -0.1700

Data Index: 369710 (C4 (Web Text))

Max Activating Token Index: 843

Text #3

Max Range: 5.2864. Min Range: -5.2864

Max Act: 5.0676. Min Act: -0.1700

Data Index: 1340019 (C4 (Web Text))

Max Activating Token Index: 737

Text #4

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.9739. Min Act: -0.1700

Data Index: 519819 (C4 (Web Text))

Max Activating Token Index: 855

Text #5

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.9739. Min Act: -0.1700

Data Index: 636013 (C4 (Web Text))

Max Activating Token Index: 863

Text #6

Max Range: 5.2864. Min Range: -5.2864

Max Act: 5.0051. Min Act: -0.1700

Data Index: 1252391 (C4 (Web Text))

Max Activating Token Index: 259

Text #7

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.9426. Min Act: -0.1700

Data Index: 610929 (C4 (Web Text))

Max Activating Token Index: 999

Text #8

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8801. Min Act: -0.1700

Data Index: 800239 (C4 (Web Text))

Max Activating Token Index: 597

Text #9

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.9739. Min Act: -0.1700

Data Index: 750463 (C4 (Web Text))

Max Activating Token Index: 931

Text #10

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.9426. Min Act: -0.1700

Data Index: 697063 (C4 (Web Text))

Max Activating Token Index: 307

Text #11

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8801. Min Act: -0.1700

Data Index: 732435 (C4 (Web Text))

Max Activating Token Index: 83

Text #12

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8801. Min Act: -0.1700

Data Index: 849924 (C4 (Web Text))

Max Activating Token Index: 885

Text #13

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.9426. Min Act: -0.1700

Data Index: 1273167 (C4 (Web Text))

Max Activating Token Index: 852

Text #14

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8801. Min Act: -0.1700

Data Index: 1063922 (C4 (Web Text))

Max Activating Token Index: 939

Text #15

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.7864. Min Act: -0.1700

Data Index: 729969 (C4 (Web Text))

Max Activating Token Index: 871

Text #16

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8489. Min Act: -0.1700

Data Index: 1340353 (C4 (Web Text))

Max Activating Token Index: 223

Text #17

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8489. Min Act: -0.1700

Data Index: 271285 (C4 (Web Text))

Max Activating Token Index: 254

Text #18

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.8489. Min Act: -0.1700

Data Index: 1154500 (C4 (Web Text))

Max Activating Token Index: 730

Text #19

Max Range: 5.2864. Min Range: -5.2864

Max Act: 4.7864. Min Act: -0.1700

Data Index: 432621 (C4 (Web Text))

Max Activating Token Index: 470