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

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

Neuron 1981 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: 3.1237. Min Range: -3.1237

Max Act: 3.1079. Min Act: -0.1700

Data Index: 487823 (C4 (Web Text))

Max Activating Token Index: 686

Text #1

Max Range: 3.1237. Min Range: -3.1237

Max Act: 3.1237. Min Act: -0.1700

Data Index: 397993 (C4 (Web Text))

Max Activating Token Index: 454

Text #2

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9816. Min Act: -0.1700

Data Index: 263505 (C4 (Web Text))

Max Activating Token Index: 973

Text #3

Max Range: 3.1237. Min Range: -3.1237

Max Act: 3.0606. Min Act: -0.1700

Data Index: 253643 (C4 (Web Text))

Max Activating Token Index: 667

Text #4

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9974. Min Act: -0.1700

Data Index: 554 (C4 (Web Text))

Max Activating Token Index: 779

Text #5

Max Range: 3.1237. Min Range: -3.1237

Max Act: 3.0132. Min Act: -0.1700

Data Index: 886446 (C4 (Web Text))

Max Activating Token Index: 207

Text #6

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9816. Min Act: -0.1700

Data Index: 1318876 (C4 (Web Text))

Max Activating Token Index: 582

Text #7

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9974. Min Act: -0.1700

Data Index: 1169756 (C4 (Web Text))

Max Activating Token Index: 37

Text #8

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9341. Min Act: -0.1700

Data Index: 19791 (C4 (Web Text))

Max Activating Token Index: 660

Text #9

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9024. Min Act: -0.1700

Data Index: 1340524 (C4 (Web Text))

Max Activating Token Index: 94

Text #10

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9024. Min Act: -0.1699

Data Index: 454196 (C4 (Web Text))

Max Activating Token Index: 833

Text #11

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9024. Min Act: -0.1700

Data Index: 1262226 (C4 (Web Text))

Max Activating Token Index: 46

Text #12

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9024. Min Act: -0.1700

Data Index: 439782 (C4 (Web Text))

Max Activating Token Index: 45

Text #13

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.8707. Min Act: -0.1700

Data Index: 492677 (C4 (Web Text))

Max Activating Token Index: 201

Text #14

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9024. Min Act: -0.1700

Data Index: 1114740 (C4 (Web Text))

Max Activating Token Index: 26

Text #15

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.8865. Min Act: -0.1700

Data Index: 1121182 (C4 (Web Text))

Max Activating Token Index: 644

Text #16

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9024. Min Act: -0.1700

Data Index: 521746 (C4 (Web Text))

Max Activating Token Index: 37

Text #17

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.8865. Min Act: -0.1700

Data Index: 1269147 (C4 (Web Text))

Max Activating Token Index: 815

Text #18

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.9183. Min Act: -0.1700

Data Index: 191184 (C4 (Web Text))

Max Activating Token Index: 471

Text #19

Max Range: 3.1237. Min Range: -3.1237

Max Act: 2.8548. Min Act: -0.1700

Data Index: 331893 (C4 (Web Text))

Max Activating Token Index: 69