Model: GELU Model: 1 Layers, 2048 Neurons per Layer

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

Neuron 1042 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('gelu-1l')

Click the toggle to see the full text!

Text #0

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.2417. Min Act: -0.1700

Data Index: 1540724 (Python Code)

Max Activating Token Index: 204

Text #1

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.2275. Min Act: -0.1699

Data Index: 328000 (C4 (Web Text))

Max Activating Token Index: 249

Text #2

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.1097. Min Act: -0.1699

Data Index: 1215015 (C4 (Web Text))

Max Activating Token Index: 32

Text #3

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.1032. Min Act: -0.1537

Data Index: 1374483 (Python Code)

Max Activating Token Index: 652

Text #4

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.0293. Min Act: -0.1616

Data Index: 1179932 (C4 (Web Text))

Max Activating Token Index: 559

Text #5

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.0155. Min Act: -0.1700

Data Index: 556851 (C4 (Web Text))

Max Activating Token Index: 160

Text #6

Max Range: 4.2417. Min Range: -4.2417

Max Act: 4.0042. Min Act: -0.1693

Data Index: 1614285 (Python Code)

Max Activating Token Index: 303

Text #7

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9997. Min Act: -0.1700

Data Index: 171008 (C4 (Web Text))

Max Activating Token Index: 710

Text #8

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9784. Min Act: -0.1696

Data Index: 1468395 (Python Code)

Max Activating Token Index: 602

Text #9

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9867. Min Act: -0.1694

Data Index: 1436478 (Python Code)

Max Activating Token Index: 627

Text #10

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9490. Min Act: -0.1664

Data Index: 107230 (C4 (Web Text))

Max Activating Token Index: 647

Text #11

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9338. Min Act: -0.1700

Data Index: 887921 (C4 (Web Text))

Max Activating Token Index: 463

Text #12

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9186. Min Act: -0.1700

Data Index: 1483831 (Python Code)

Max Activating Token Index: 791

Text #13

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9167. Min Act: -0.1662

Data Index: 1526074 (Python Code)

Max Activating Token Index: 937

Text #14

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9129. Min Act: -0.1700

Data Index: 1392956 (Python Code)

Max Activating Token Index: 311

Text #15

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.8840. Min Act: -0.1700

Data Index: 1603818 (Python Code)

Max Activating Token Index: 953

Text #16

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9132. Min Act: -0.1680

Data Index: 717775 (C4 (Web Text))

Max Activating Token Index: 186

Text #17

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.9001. Min Act: -0.1673

Data Index: 813940 (C4 (Web Text))

Max Activating Token Index: 1018

Text #18

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.8886. Min Act: -0.1698

Data Index: 1652255 (Python Code)

Max Activating Token Index: 885

Text #19

Max Range: 4.2417. Min Range: -4.2417

Max Act: 3.8793. Min Act: -0.1700

Data Index: 1425401 (Python Code)

Max Activating Token Index: 479