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

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

Neuron 1746 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: 9.1838. Min Range: -9.1838

Max Act: 9.1838. Min Act: -0.1700

Data Index: 887573 (C4 (Web Text))

Max Activating Token Index: 259

Text #1

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.8713. Min Act: -0.1700

Data Index: 579710 (C4 (Web Text))

Max Activating Token Index: 158

Text #2

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.6526. Min Act: -0.1700

Data Index: 705930 (C4 (Web Text))

Max Activating Token Index: 1014

Text #3

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.7776. Min Act: -0.1700

Data Index: 1197842 (C4 (Web Text))

Max Activating Token Index: 233

Text #4

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.9026. Min Act: -0.1700

Data Index: 1254912 (C4 (Web Text))

Max Activating Token Index: 248

Text #5

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.7151. Min Act: -0.1700

Data Index: 1003419 (C4 (Web Text))

Max Activating Token Index: 623

Text #6

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.7463. Min Act: -0.1700

Data Index: 1404016 (Python Code)

Max Activating Token Index: 722

Text #7

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.6213. Min Act: -0.1700

Data Index: 380058 (C4 (Web Text))

Max Activating Token Index: 468

Text #8

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.5901. Min Act: -0.1700

Data Index: 806453 (C4 (Web Text))

Max Activating Token Index: 259

Text #9

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.5588. Min Act: -0.1700

Data Index: 631999 (C4 (Web Text))

Max Activating Token Index: 591

Text #10

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.6213. Min Act: -0.1700

Data Index: 596188 (C4 (Web Text))

Max Activating Token Index: 442

Text #11

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.5276. Min Act: -0.1700

Data Index: 1154948 (C4 (Web Text))

Max Activating Token Index: 111

Text #12

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.3713. Min Act: -0.1700

Data Index: 3455 (C4 (Web Text))

Max Activating Token Index: 903

Text #13

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.4963. Min Act: -0.1700

Data Index: 1074248 (C4 (Web Text))

Max Activating Token Index: 164

Text #14

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.2776. Min Act: -0.1700

Data Index: 555219 (C4 (Web Text))

Max Activating Token Index: 117

Text #15

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.3088. Min Act: -0.1700

Data Index: 1242543 (C4 (Web Text))

Max Activating Token Index: 314

Text #16

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.3088. Min Act: -0.1700

Data Index: 1634244 (Python Code)

Max Activating Token Index: 817

Text #17

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.3401. Min Act: -0.1700

Data Index: 1384459 (Python Code)

Max Activating Token Index: 234

Text #18

Max Range: 9.1838. Min Range: -9.1838

Max Act: 8.3088. Min Act: -0.1700

Data Index: 1319412 (C4 (Web Text))

Max Activating Token Index: 321

Text #19

Max Range: 9.1838. Min Range: -9.1838

Max Act: 7.9651. Min Act: -0.1700

Data Index: 427865 (C4 (Web Text))

Max Activating Token Index: 539