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

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

Neuron 775 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: 5.4991. Min Range: -5.4991

Max Act: 5.4991. Min Act: -0.1700

Data Index: 593179 (C4 (Web Text))

Max Activating Token Index: 610

Text #1

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.3116. Min Act: -0.1700

Data Index: 1354731 (C4 (Web Text))

Max Activating Token Index: 346

Text #2

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.2178. Min Act: -0.1700

Data Index: 1112256 (C4 (Web Text))

Max Activating Token Index: 675

Text #3

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.1866. Min Act: -0.1700

Data Index: 332372 (C4 (Web Text))

Max Activating Token Index: 100

Text #4

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.1553. Min Act: -0.1700

Data Index: 3383 (C4 (Web Text))

Max Activating Token Index: 629

Text #5

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.0616. Min Act: -0.1700

Data Index: 585890 (C4 (Web Text))

Max Activating Token Index: 812

Text #6

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.0928. Min Act: -0.1700

Data Index: 1099745 (C4 (Web Text))

Max Activating Token Index: 506

Text #7

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.0303. Min Act: -0.1700

Data Index: 423792 (C4 (Web Text))

Max Activating Token Index: 801

Text #8

Max Range: 5.4991. Min Range: -5.4991

Max Act: 5.0303. Min Act: -0.1700

Data Index: 1244889 (C4 (Web Text))

Max Activating Token Index: 579

Text #9

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9366. Min Act: -0.1700

Data Index: 959194 (C4 (Web Text))

Max Activating Token Index: 510

Text #10

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.8741. Min Act: -0.1700

Data Index: 401159 (C4 (Web Text))

Max Activating Token Index: 337

Text #11

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9366. Min Act: -0.1700

Data Index: 1132246 (C4 (Web Text))

Max Activating Token Index: 315

Text #12

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9678. Min Act: -0.1700

Data Index: 1244408 (C4 (Web Text))

Max Activating Token Index: 753

Text #13

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9366. Min Act: -0.1700

Data Index: 1035238 (C4 (Web Text))

Max Activating Token Index: 886

Text #14

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9053. Min Act: -0.1700

Data Index: 432060 (C4 (Web Text))

Max Activating Token Index: 827

Text #15

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9366. Min Act: -0.1700

Data Index: 84498 (C4 (Web Text))

Max Activating Token Index: 684

Text #16

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9053. Min Act: -0.1700

Data Index: 827448 (C4 (Web Text))

Max Activating Token Index: 397

Text #17

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9053. Min Act: -0.1700

Data Index: 448025 (C4 (Web Text))

Max Activating Token Index: 360

Text #18

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.9053. Min Act: -0.1700

Data Index: 976447 (C4 (Web Text))

Max Activating Token Index: 53

Text #19

Max Range: 5.4991. Min Range: -5.4991

Max Act: 4.8741. Min Act: -0.1700

Data Index: 750737 (C4 (Web Text))

Max Activating Token Index: 531