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

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

Neuron 1535 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: 3.2577. Min Range: -3.2577

Max Act: 3.2577. Min Act: -0.1700

Data Index: 105400 (C4 (Web Text))

Max Activating Token Index: 646

Text #1

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.2262. Min Act: -0.1700

Data Index: 1057676 (C4 (Web Text))

Max Activating Token Index: 638

Text #2

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.2262. Min Act: -0.1700

Data Index: 52880 (C4 (Web Text))

Max Activating Token Index: 758

Text #3

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1476. Min Act: -0.1700

Data Index: 93180 (C4 (Web Text))

Max Activating Token Index: 6

Text #4

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1948. Min Act: -0.1700

Data Index: 182167 (C4 (Web Text))

Max Activating Token Index: 663

Text #5

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1476. Min Act: -0.1700

Data Index: 1550644 (Python Code)

Max Activating Token Index: 1023

Text #6

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1318. Min Act: -0.1700

Data Index: 348715 (C4 (Web Text))

Max Activating Token Index: 578

Text #7

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1318. Min Act: -0.1700

Data Index: 1345100 (C4 (Web Text))

Max Activating Token Index: 598

Text #8

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1161. Min Act: -0.1700

Data Index: 147619 (C4 (Web Text))

Max Activating Token Index: 88

Text #9

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0687. Min Act: -0.1700

Data Index: 714189 (C4 (Web Text))

Max Activating Token Index: 772

Text #10

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0845. Min Act: -0.1700

Data Index: 26228 (C4 (Web Text))

Max Activating Token Index: 595

Text #11

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1161. Min Act: -0.1700

Data Index: 478798 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1003. Min Act: -0.1700

Data Index: 740777 (C4 (Web Text))

Max Activating Token Index: 106

Text #13

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.1003. Min Act: -0.1700

Data Index: 220619 (C4 (Web Text))

Max Activating Token Index: 585

Text #14

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0687. Min Act: -0.1700

Data Index: 945652 (C4 (Web Text))

Max Activating Token Index: 905

Text #15

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0845. Min Act: -0.1700

Data Index: 243675 (C4 (Web Text))

Max Activating Token Index: 871

Text #16

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0687. Min Act: -0.1700

Data Index: 646498 (C4 (Web Text))

Max Activating Token Index: 901

Text #17

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0372. Min Act: -0.1700

Data Index: 1568233 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0372. Min Act: -0.1700

Data Index: 704514 (C4 (Web Text))

Max Activating Token Index: 714

Text #19

Max Range: 3.2577. Min Range: -3.2577

Max Act: 3.0687. Min Act: -0.1700

Data Index: 912661 (C4 (Web Text))

Max Activating Token Index: 909