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

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

Neuron 1309 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: 4.1744. Min Range: -4.1744

Max Act: 4.1744. Min Act: -0.1700

Data Index: 1597688 (Python Code)

Max Activating Token Index: 639

Text #1

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.8930. Min Act: -0.1700

Data Index: 27555 (C4 (Web Text))

Max Activating Token Index: 754

Text #2

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.8930. Min Act: -0.1700

Data Index: 109165 (C4 (Web Text))

Max Activating Token Index: 342

Text #3

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.8930. Min Act: -0.1700

Data Index: 996872 (C4 (Web Text))

Max Activating Token Index: 394

Text #4

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.8618. Min Act: -0.1700

Data Index: 1161807 (C4 (Web Text))

Max Activating Token Index: 87

Text #5

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7679. Min Act: -0.1700

Data Index: 1626626 (Python Code)

Max Activating Token Index: 869

Text #6

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7992. Min Act: -0.1700

Data Index: 1009398 (C4 (Web Text))

Max Activating Token Index: 265

Text #7

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7992. Min Act: -0.1700

Data Index: 1221375 (C4 (Web Text))

Max Activating Token Index: 1020

Text #8

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7992. Min Act: -0.1700

Data Index: 786126 (C4 (Web Text))

Max Activating Token Index: 827

Text #9

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7679. Min Act: -0.1700

Data Index: 724965 (C4 (Web Text))

Max Activating Token Index: 810

Text #10

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7366. Min Act: -0.1700

Data Index: 1596187 (Python Code)

Max Activating Token Index: 157

Text #11

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7366. Min Act: -0.1700

Data Index: 1433633 (Python Code)

Max Activating Token Index: 939

Text #12

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7366. Min Act: -0.1700

Data Index: 307590 (C4 (Web Text))

Max Activating Token Index: 759

Text #13

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.6740. Min Act: -0.1700

Data Index: 678236 (C4 (Web Text))

Max Activating Token Index: 950

Text #14

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7053. Min Act: -0.1700

Data Index: 419424 (C4 (Web Text))

Max Activating Token Index: 120

Text #15

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7053. Min Act: -0.1700

Data Index: 282495 (C4 (Web Text))

Max Activating Token Index: 1008

Text #16

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7053. Min Act: -0.1700

Data Index: 1035156 (C4 (Web Text))

Max Activating Token Index: 208

Text #17

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7366. Min Act: -0.1700

Data Index: 605553 (C4 (Web Text))

Max Activating Token Index: 123

Text #18

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.7053. Min Act: -0.1700

Data Index: 1034322 (C4 (Web Text))

Max Activating Token Index: 522

Text #19

Max Range: 4.1744. Min Range: -4.1744

Max Act: 3.6740. Min Act: -0.1700

Data Index: 1226587 (C4 (Web Text))

Max Activating Token Index: 563