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

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

Neuron 1454 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.8434. Min Act: -0.1700

Data Index: 606302 (C4 (Web Text))

Max Activating Token Index: 811

Text #1

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.8747. Min Act: -0.1700

Data Index: 816732 (C4 (Web Text))

Max Activating Token Index: 1001

Text #2

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.7809. Min Act: -0.1700

Data Index: 989278 (C4 (Web Text))

Max Activating Token Index: 860

Text #3

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.7496. Min Act: -0.1700

Data Index: 1050501 (C4 (Web Text))

Max Activating Token Index: 346

Text #4

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.8122. Min Act: -0.1700

Data Index: 508209 (C4 (Web Text))

Max Activating Token Index: 490

Text #5

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.7809. Min Act: -0.1700

Data Index: 12669 (C4 (Web Text))

Max Activating Token Index: 577

Text #6

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.7809. Min Act: -0.1700

Data Index: 73359 (C4 (Web Text))

Max Activating Token Index: 927

Text #7

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.6557. Min Act: -0.1700

Data Index: 1518672 (Python Code)

Max Activating Token Index: 259

Text #8

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.6557. Min Act: -0.1700

Data Index: 471308 (C4 (Web Text))

Max Activating Token Index: 270

Text #9

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.6244. Min Act: -0.1700

Data Index: 1106558 (C4 (Web Text))

Max Activating Token Index: 706

Text #10

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5931. Min Act: -0.1700

Data Index: 215393 (C4 (Web Text))

Max Activating Token Index: 42

Text #11

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5931. Min Act: -0.1700

Data Index: 159107 (C4 (Web Text))

Max Activating Token Index: 316

Text #12

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5618. Min Act: -0.1700

Data Index: 1288647 (C4 (Web Text))

Max Activating Token Index: 28

Text #13

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5931. Min Act: -0.1700

Data Index: 555822 (C4 (Web Text))

Max Activating Token Index: 313

Text #14

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.6244. Min Act: -0.1700

Data Index: 251949 (C4 (Web Text))

Max Activating Token Index: 478

Text #15

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.6557. Min Act: -0.1700

Data Index: 472631 (C4 (Web Text))

Max Activating Token Index: 77

Text #16

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5618. Min Act: -0.1700

Data Index: 1496820 (Python Code)

Max Activating Token Index: 346

Text #17

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5304. Min Act: -0.1700

Data Index: 683310 (C4 (Web Text))

Max Activating Token Index: 751

Text #18

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5931. Min Act: -0.1700

Data Index: 452537 (C4 (Web Text))

Max Activating Token Index: 585

Text #19

Max Range: 3.8747. Min Range: -3.8747

Max Act: 3.5618. Min Act: -0.1700

Data Index: 905945 (C4 (Web Text))

Max Activating Token Index: 686