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

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

Neuron 39 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: 3.2590. Min Range: -3.2590

Max Act: 3.2590. Min Act: -0.1700

Data Index: 797028 (C4 (Web Text))

Max Activating Token Index: 899

Text #1

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.1492. Min Act: -0.1700

Data Index: 838095 (C4 (Web Text))

Max Activating Token Index: 805

Text #2

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.1121. Min Act: -0.1700

Data Index: 1580345 (Python Code)

Max Activating Token Index: 464

Text #3

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0899. Min Act: -0.1700

Data Index: 897682 (C4 (Web Text))

Max Activating Token Index: 766

Text #4

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0796. Min Act: -0.1700

Data Index: 668688 (C4 (Web Text))

Max Activating Token Index: 571

Text #5

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.1006. Min Act: -0.1700

Data Index: 1330559 (C4 (Web Text))

Max Activating Token Index: 1020

Text #6

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0674. Min Act: -0.1700

Data Index: 137813 (C4 (Web Text))

Max Activating Token Index: 3

Text #7

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0494. Min Act: -0.1700

Data Index: 891870 (C4 (Web Text))

Max Activating Token Index: 590

Text #8

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0493. Min Act: -0.1700

Data Index: 1391219 (Python Code)

Max Activating Token Index: 681

Text #9

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0357. Min Act: -0.1700

Data Index: 1303191 (C4 (Web Text))

Max Activating Token Index: 283

Text #10

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0118. Min Act: -0.1700

Data Index: 1247010 (C4 (Web Text))

Max Activating Token Index: 633

Text #11

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0073. Min Act: -0.1700

Data Index: 1552492 (Python Code)

Max Activating Token Index: 50

Text #12

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0176. Min Act: -0.1700

Data Index: 1456792 (Python Code)

Max Activating Token Index: 7

Text #13

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0089. Min Act: -0.1700

Data Index: 598487 (C4 (Web Text))

Max Activating Token Index: 112

Text #14

Max Range: 3.2590. Min Range: -3.2590

Max Act: 2.9826. Min Act: -0.1700

Data Index: 190451 (C4 (Web Text))

Max Activating Token Index: 772

Text #15

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0377. Min Act: -0.1700

Data Index: 1528510 (Python Code)

Max Activating Token Index: 558

Text #16

Max Range: 3.2590. Min Range: -3.2590

Max Act: 3.0014. Min Act: -0.1700

Data Index: 1466784 (Python Code)

Max Activating Token Index: 200

Text #17

Max Range: 3.2590. Min Range: -3.2590

Max Act: 2.9757. Min Act: -0.1700

Data Index: 218851 (C4 (Web Text))

Max Activating Token Index: 482

Text #18

Max Range: 3.2590. Min Range: -3.2590

Max Act: 2.9694. Min Act: -0.1700

Data Index: 614582 (C4 (Web Text))

Max Activating Token Index: 107

Text #19

Max Range: 3.2590. Min Range: -3.2590

Max Act: 2.9963. Min Act: -0.1700

Data Index: 834930 (C4 (Web Text))

Max Activating Token Index: 11