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

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

Neuron 32 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: 3.7372. Min Range: -3.7372

Max Act: 3.7372. Min Act: -0.1700

Data Index: 1559744 (Python Code)

Max Activating Token Index: 676

Text #1

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.6433. Min Act: -0.1700

Data Index: 1436325 (Python Code)

Max Activating Token Index: 606

Text #2

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.6120. Min Act: -0.1700

Data Index: 1625835 (Python Code)

Max Activating Token Index: 577

Text #3

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.5493. Min Act: -0.1699

Data Index: 1086696 (C4 (Web Text))

Max Activating Token Index: 314

Text #4

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.5180. Min Act: -0.1700

Data Index: 212081 (C4 (Web Text))

Max Activating Token Index: 373

Text #5

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.5180. Min Act: -0.1700

Data Index: 1540671 (Python Code)

Max Activating Token Index: 769

Text #6

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.5180. Min Act: -0.1700

Data Index: 1567591 (Python Code)

Max Activating Token Index: 882

Text #7

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.5180. Min Act: -0.1700

Data Index: 331869 (C4 (Web Text))

Max Activating Token Index: 247

Text #8

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4553. Min Act: -0.1700

Data Index: 1395543 (Python Code)

Max Activating Token Index: 937

Text #9

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4867. Min Act: -0.1700

Data Index: 1437799 (Python Code)

Max Activating Token Index: 357

Text #10

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4867. Min Act: -0.1700

Data Index: 393425 (C4 (Web Text))

Max Activating Token Index: 361

Text #11

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.5180. Min Act: -0.1700

Data Index: 92873 (C4 (Web Text))

Max Activating Token Index: 510

Text #12

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4553. Min Act: -0.1700

Data Index: 1600487 (Python Code)

Max Activating Token Index: 955

Text #13

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4553. Min Act: -0.1700

Data Index: 287228 (C4 (Web Text))

Max Activating Token Index: 110

Text #14

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4240. Min Act: -0.1700

Data Index: 1190813 (C4 (Web Text))

Max Activating Token Index: 471

Text #15

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4553. Min Act: -0.1700

Data Index: 352183 (C4 (Web Text))

Max Activating Token Index: 336

Text #16

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4240. Min Act: -0.1700

Data Index: 261009 (C4 (Web Text))

Max Activating Token Index: 271

Text #17

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4240. Min Act: -0.1700

Data Index: 1282745 (C4 (Web Text))

Max Activating Token Index: 322

Text #18

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4240. Min Act: -0.1700

Data Index: 504138 (C4 (Web Text))

Max Activating Token Index: 523

Text #19

Max Range: 3.7372. Min Range: -3.7372

Max Act: 3.4240. Min Act: -0.1700

Data Index: 1379848 (Python Code)

Max Activating Token Index: 578