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

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

Neuron 308 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.0560. Min Range: -5.0560

Max Act: 5.0560. Min Act: -0.1700

Data Index: 688748 (C4 (Web Text))

Max Activating Token Index: 764

Text #1

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.9623. Min Act: -0.1700

Data Index: 77879 (C4 (Web Text))

Max Activating Token Index: 511

Text #2

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.9466. Min Act: -0.1700

Data Index: 416737 (C4 (Web Text))

Max Activating Token Index: 847

Text #3

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.8529. Min Act: -0.1700

Data Index: 774736 (C4 (Web Text))

Max Activating Token Index: 783

Text #4

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.8841. Min Act: -0.1700

Data Index: 1349613 (C4 (Web Text))

Max Activating Token Index: 162

Text #5

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.9466. Min Act: -0.1699

Data Index: 739074 (C4 (Web Text))

Max Activating Token Index: 242

Text #6

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.8841. Min Act: -0.1700

Data Index: 716863 (C4 (Web Text))

Max Activating Token Index: 913

Text #7

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.8685. Min Act: -0.1700

Data Index: 1035134 (C4 (Web Text))

Max Activating Token Index: 265

Text #8

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.8373. Min Act: -0.1700

Data Index: 1139010 (C4 (Web Text))

Max Activating Token Index: 60

Text #9

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.8529. Min Act: -0.1700

Data Index: 430723 (C4 (Web Text))

Max Activating Token Index: 641

Text #10

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.7904. Min Act: -0.1700

Data Index: 645963 (C4 (Web Text))

Max Activating Token Index: 352

Text #11

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.7904. Min Act: -0.1700

Data Index: 1163812 (C4 (Web Text))

Max Activating Token Index: 33

Text #12

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.7904. Min Act: -0.1700

Data Index: 337041 (C4 (Web Text))

Max Activating Token Index: 940

Text #13

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.7279. Min Act: -0.1700

Data Index: 1120963 (C4 (Web Text))

Max Activating Token Index: 848

Text #14

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.7279. Min Act: -0.1700

Data Index: 720877 (C4 (Web Text))

Max Activating Token Index: 932

Text #15

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.7123. Min Act: -0.1700

Data Index: 646877 (C4 (Web Text))

Max Activating Token Index: 113

Text #16

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.6498. Min Act: -0.1700

Data Index: 1277713 (C4 (Web Text))

Max Activating Token Index: 932

Text #17

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.6654. Min Act: -0.1700

Data Index: 1055948 (C4 (Web Text))

Max Activating Token Index: 475

Text #18

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.6185. Min Act: -0.1699

Data Index: 1016902 (C4 (Web Text))

Max Activating Token Index: 372

Text #19

Max Range: 5.0560. Min Range: -5.0560

Max Act: 4.6029. Min Act: -0.1700

Data Index: 631089 (C4 (Web Text))

Max Activating Token Index: 675