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

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

Neuron 1828 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: 6.0088. Min Range: -6.0088

Max Act: 6.0088. Min Act: -0.1700

Data Index: 82612 (C4 (Web Text))

Max Activating Token Index: 694

Text #1

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.8525. Min Act: -0.1700

Data Index: 965658 (C4 (Web Text))

Max Activating Token Index: 502

Text #2

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.8213. Min Act: -0.1700

Data Index: 140580 (C4 (Web Text))

Max Activating Token Index: 273

Text #3

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.8213. Min Act: -0.1700

Data Index: 1031920 (C4 (Web Text))

Max Activating Token Index: 936

Text #4

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.7900. Min Act: -0.1700

Data Index: 737792 (C4 (Web Text))

Max Activating Token Index: 20

Text #5

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.7588. Min Act: -0.1700

Data Index: 1094451 (C4 (Web Text))

Max Activating Token Index: 726

Text #6

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.7588. Min Act: -0.1700

Data Index: 1332936 (C4 (Web Text))

Max Activating Token Index: 294

Text #7

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.6963. Min Act: -0.1700

Data Index: 70511 (C4 (Web Text))

Max Activating Token Index: 255

Text #8

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.6025. Min Act: -0.1700

Data Index: 1053526 (C4 (Web Text))

Max Activating Token Index: 779

Text #9

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.6025. Min Act: -0.1700

Data Index: 7487 (C4 (Web Text))

Max Activating Token Index: 940

Text #10

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.5713. Min Act: -0.1700

Data Index: 478877 (C4 (Web Text))

Max Activating Token Index: 768

Text #11

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.6025. Min Act: -0.1700

Data Index: 528282 (C4 (Web Text))

Max Activating Token Index: 833

Text #12

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.5088. Min Act: -0.1700

Data Index: 105664 (C4 (Web Text))

Max Activating Token Index: 70

Text #13

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.4775. Min Act: -0.1700

Data Index: 1293522 (C4 (Web Text))

Max Activating Token Index: 453

Text #14

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.5088. Min Act: -0.1700

Data Index: 1090622 (C4 (Web Text))

Max Activating Token Index: 604

Text #15

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.5400. Min Act: -0.1700

Data Index: 734740 (C4 (Web Text))

Max Activating Token Index: 888

Text #16

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.5088. Min Act: -0.1700

Data Index: 795306 (C4 (Web Text))

Max Activating Token Index: 33

Text #17

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.4463. Min Act: -0.1700

Data Index: 267724 (C4 (Web Text))

Max Activating Token Index: 350

Text #18

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.4463. Min Act: -0.1700

Data Index: 973464 (C4 (Web Text))

Max Activating Token Index: 858

Text #19

Max Range: 6.0088. Min Range: -6.0088

Max Act: 5.4775. Min Act: -0.1700

Data Index: 482581 (C4 (Web Text))

Max Activating Token Index: 656