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

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

Neuron 544 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.9400. Min Act: -0.1700

Data Index: 645813 (C4 (Web Text))

Max Activating Token Index: 99

Text #1

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.9713. Min Act: -0.1700

Data Index: 971032 (C4 (Web Text))

Max Activating Token Index: 430

Text #2

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.9400. Min Act: -0.1700

Data Index: 224763 (C4 (Web Text))

Max Activating Token Index: 700

Text #3

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8775. Min Act: -0.1700

Data Index: 662510 (C4 (Web Text))

Max Activating Token Index: 476

Text #4

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8775. Min Act: -0.1700

Data Index: 85018 (C4 (Web Text))

Max Activating Token Index: 555

Text #5

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8462. Min Act: -0.1700

Data Index: 801652 (C4 (Web Text))

Max Activating Token Index: 493

Text #6

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8462. Min Act: -0.1700

Data Index: 965677 (C4 (Web Text))

Max Activating Token Index: 299

Text #7

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8462. Min Act: -0.1700

Data Index: 532798 (C4 (Web Text))

Max Activating Token Index: 686

Text #8

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 1310996 (C4 (Web Text))

Max Activating Token Index: 10

Text #9

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8149. Min Act: -0.1700

Data Index: 916213 (C4 (Web Text))

Max Activating Token Index: 737

Text #10

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 1172716 (C4 (Web Text))

Max Activating Token Index: 797

Text #11

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8149. Min Act: -0.1700

Data Index: 662064 (C4 (Web Text))

Max Activating Token Index: 242

Text #12

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 680964 (C4 (Web Text))

Max Activating Token Index: 935

Text #13

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 650058 (C4 (Web Text))

Max Activating Token Index: 459

Text #14

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 617749 (C4 (Web Text))

Max Activating Token Index: 80

Text #15

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 489780 (C4 (Web Text))

Max Activating Token Index: 716

Text #16

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8149. Min Act: -0.1700

Data Index: 519528 (C4 (Web Text))

Max Activating Token Index: 642

Text #17

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7836. Min Act: -0.1700

Data Index: 1096007 (C4 (Web Text))

Max Activating Token Index: 564

Text #18

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.7523. Min Act: -0.1700

Data Index: 872539 (C4 (Web Text))

Max Activating Token Index: 482

Text #19

Max Range: 3.9713. Min Range: -3.9713

Max Act: 3.8149. Min Act: -0.1700

Data Index: 1138973 (C4 (Web Text))

Max Activating Token Index: 995