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

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

Neuron 132 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.4903. Min Act: -0.1700

Data Index: 1223100 (C4 (Web Text))

Max Activating Token Index: 411

Text #1

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.2559. Min Act: -0.1700

Data Index: 911126 (C4 (Web Text))

Max Activating Token Index: 835

Text #2

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.2247. Min Act: -0.1700

Data Index: 982733 (C4 (Web Text))

Max Activating Token Index: 740

Text #3

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.1465. Min Act: -0.1700

Data Index: 1149289 (C4 (Web Text))

Max Activating Token Index: 932

Text #4

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.1622. Min Act: -0.1700

Data Index: 223393 (C4 (Web Text))

Max Activating Token Index: 184

Text #5

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.1778. Min Act: -0.1700

Data Index: 80898 (C4 (Web Text))

Max Activating Token Index: 684

Text #6

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.1153. Min Act: -0.1700

Data Index: 369290 (C4 (Web Text))

Max Activating Token Index: 378

Text #7

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.1309. Min Act: -0.1700

Data Index: 896819 (C4 (Web Text))

Max Activating Token Index: 499

Text #8

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.1153. Min Act: -0.1700

Data Index: 1114024 (C4 (Web Text))

Max Activating Token Index: 341

Text #9

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.0997. Min Act: -0.1700

Data Index: 363775 (C4 (Web Text))

Max Activating Token Index: 280

Text #10

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9590. Min Act: -0.1700

Data Index: 1135661 (C4 (Web Text))

Max Activating Token Index: 446

Text #11

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.0372. Min Act: -0.1700

Data Index: 1352047 (C4 (Web Text))

Max Activating Token Index: 216

Text #12

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9903. Min Act: -0.1700

Data Index: 371620 (C4 (Web Text))

Max Activating Token Index: 350

Text #13

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9590. Min Act: -0.1700

Data Index: 1243241 (C4 (Web Text))

Max Activating Token Index: 396

Text #14

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.0215. Min Act: -0.1700

Data Index: 989092 (C4 (Web Text))

Max Activating Token Index: 555

Text #15

Max Range: 5.4903. Min Range: -5.4903

Max Act: 5.0215. Min Act: -0.1700

Data Index: 843982 (C4 (Web Text))

Max Activating Token Index: 951

Text #16

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9434. Min Act: -0.1700

Data Index: 648607 (C4 (Web Text))

Max Activating Token Index: 737

Text #17

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9903. Min Act: -0.1700

Data Index: 921476 (C4 (Web Text))

Max Activating Token Index: 383

Text #18

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9903. Min Act: -0.1700

Data Index: 654244 (C4 (Web Text))

Max Activating Token Index: 602

Text #19

Max Range: 5.4903. Min Range: -5.4903

Max Act: 4.9434. Min Act: -0.1700

Data Index: 641423 (C4 (Web Text))

Max Activating Token Index: 358