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

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

Neuron 355 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.1968. Min Act: -0.1700

Data Index: 1062662 (C4 (Web Text))

Max Activating Token Index: 73

Text #1

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.1810. Min Act: -0.1700

Data Index: 576762 (C4 (Web Text))

Max Activating Token Index: 1012

Text #2

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.1338. Min Act: -0.1700

Data Index: 1435878 (Python Code)

Max Activating Token Index: 516

Text #3

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.1180. Min Act: -0.1700

Data Index: 572050 (C4 (Web Text))

Max Activating Token Index: 356

Text #4

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.1023. Min Act: -0.1700

Data Index: 1443331 (Python Code)

Max Activating Token Index: 5

Text #5

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0865. Min Act: -0.1700

Data Index: 760585 (C4 (Web Text))

Max Activating Token Index: 393

Text #6

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0865. Min Act: -0.1700

Data Index: 1495501 (Python Code)

Max Activating Token Index: 520

Text #7

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0707. Min Act: -0.1700

Data Index: 227455 (C4 (Web Text))

Max Activating Token Index: 704

Text #8

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0707. Min Act: -0.1700

Data Index: 315331 (C4 (Web Text))

Max Activating Token Index: 395

Text #9

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0707. Min Act: -0.1700

Data Index: 827054 (C4 (Web Text))

Max Activating Token Index: 392

Text #10

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0549. Min Act: -0.1700

Data Index: 444367 (C4 (Web Text))

Max Activating Token Index: 616

Text #11

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0549. Min Act: -0.1700

Data Index: 100448 (C4 (Web Text))

Max Activating Token Index: 755

Text #12

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0549. Min Act: -0.1700

Data Index: 1284225 (C4 (Web Text))

Max Activating Token Index: 477

Text #13

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0391. Min Act: -0.1700

Data Index: 1151512 (C4 (Web Text))

Max Activating Token Index: 854

Text #14

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0549. Min Act: -0.1700

Data Index: 1313763 (C4 (Web Text))

Max Activating Token Index: 618

Text #15

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0391. Min Act: -0.1700

Data Index: 1448659 (Python Code)

Max Activating Token Index: 719

Text #16

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0391. Min Act: -0.1700

Data Index: 799682 (C4 (Web Text))

Max Activating Token Index: 246

Text #17

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0233. Min Act: -0.1700

Data Index: 290011 (C4 (Web Text))

Max Activating Token Index: 354

Text #18

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0075. Min Act: -0.1700

Data Index: 992953 (C4 (Web Text))

Max Activating Token Index: 587

Text #19

Max Range: 3.1968. Min Range: -3.1968

Max Act: 3.0075. Min Act: -0.1700

Data Index: 148608 (C4 (Web Text))

Max Activating Token Index: 157