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

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

Neuron 1104 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: 3.8377. Min Range: -3.8377

Max Act: 3.8377. Min Act: -0.1700

Data Index: 1437788 (Python Code)

Max Activating Token Index: 34

Text #1

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.6947. Min Act: -0.1700

Data Index: 579137 (C4 (Web Text))

Max Activating Token Index: 43

Text #2

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.6372. Min Act: -0.1700

Data Index: 34285 (C4 (Web Text))

Max Activating Token Index: 59

Text #3

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.6268. Min Act: -0.1700

Data Index: 383878 (C4 (Web Text))

Max Activating Token Index: 262

Text #4

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.6047. Min Act: -0.1700

Data Index: 81015 (C4 (Web Text))

Max Activating Token Index: 691

Text #5

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.5960. Min Act: -0.1700

Data Index: 1548703 (Python Code)

Max Activating Token Index: 28

Text #6

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.5582. Min Act: -0.1700

Data Index: 1231377 (C4 (Web Text))

Max Activating Token Index: 730

Text #7

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.5520. Min Act: -0.1700

Data Index: 535056 (C4 (Web Text))

Max Activating Token Index: 511

Text #8

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.5368. Min Act: -0.1700

Data Index: 1015881 (C4 (Web Text))

Max Activating Token Index: 332

Text #9

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.5232. Min Act: -0.1700

Data Index: 96117 (C4 (Web Text))

Max Activating Token Index: 719

Text #10

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.5018. Min Act: -0.1700

Data Index: 70505 (C4 (Web Text))

Max Activating Token Index: 147

Text #11

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4993. Min Act: -0.1700

Data Index: 122507 (C4 (Web Text))

Max Activating Token Index: 689

Text #12

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4772. Min Act: -0.1700

Data Index: 1335777 (C4 (Web Text))

Max Activating Token Index: 797

Text #13

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4838. Min Act: -0.1700

Data Index: 1227561 (C4 (Web Text))

Max Activating Token Index: 8

Text #14

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4685. Min Act: -0.1700

Data Index: 519264 (C4 (Web Text))

Max Activating Token Index: 218

Text #15

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4575. Min Act: -0.1700

Data Index: 797147 (C4 (Web Text))

Max Activating Token Index: 921

Text #16

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4653. Min Act: -0.1700

Data Index: 956046 (C4 (Web Text))

Max Activating Token Index: 780

Text #17

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4344. Min Act: -0.1700

Data Index: 1546589 (Python Code)

Max Activating Token Index: 246

Text #18

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4386. Min Act: -0.1700

Data Index: 1430698 (Python Code)

Max Activating Token Index: 226

Text #19

Max Range: 3.8377. Min Range: -3.8377

Max Act: 3.4072. Min Act: -0.1700

Data Index: 1447710 (Python Code)

Max Activating Token Index: 737