Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1214 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.8643. Min Act: -0.1700

Data Index: 1331946 (C4 (Web Text))

Max Activating Token Index: 392

Text #1

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.8955. Min Act: -0.1700

Data Index: 40553 (C4 (Web Text))

Max Activating Token Index: 327

Text #2

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.8017. Min Act: -0.1700

Data Index: 997385 (C4 (Web Text))

Max Activating Token Index: 891

Text #3

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.7704. Min Act: -0.1700

Data Index: 568826 (C4 (Web Text))

Max Activating Token Index: 663

Text #4

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.7704. Min Act: -0.1700

Data Index: 314491 (C4 (Web Text))

Max Activating Token Index: 848

Text #5

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.7391. Min Act: -0.1700

Data Index: 704161 (C4 (Web Text))

Max Activating Token Index: 606

Text #6

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.7078. Min Act: -0.1700

Data Index: 703605 (C4 (Web Text))

Max Activating Token Index: 1020

Text #7

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.7548. Min Act: -0.1700

Data Index: 1159450 (C4 (Web Text))

Max Activating Token Index: 214

Text #8

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.7078. Min Act: -0.1700

Data Index: 1199145 (C4 (Web Text))

Max Activating Token Index: 988

Text #9

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.5826. Min Act: -0.1700

Data Index: 796067 (C4 (Web Text))

Max Activating Token Index: 833

Text #10

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6765. Min Act: -0.1700

Data Index: 1348436 (C4 (Web Text))

Max Activating Token Index: 712

Text #11

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6452. Min Act: -0.1700

Data Index: 996363 (C4 (Web Text))

Max Activating Token Index: 282

Text #12

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6452. Min Act: -0.1700

Data Index: 836333 (C4 (Web Text))

Max Activating Token Index: 442

Text #13

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.5670. Min Act: -0.1700

Data Index: 642018 (C4 (Web Text))

Max Activating Token Index: 876

Text #14

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6296. Min Act: -0.1700

Data Index: 1329785 (C4 (Web Text))

Max Activating Token Index: 128

Text #15

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6296. Min Act: -0.1700

Data Index: 635304 (C4 (Web Text))

Max Activating Token Index: 958

Text #16

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.5670. Min Act: -0.1700

Data Index: 255238 (C4 (Web Text))

Max Activating Token Index: 275

Text #17

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6139. Min Act: -0.1700

Data Index: 112901 (C4 (Web Text))

Max Activating Token Index: 662

Text #18

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.6139. Min Act: -0.1700

Data Index: 313799 (C4 (Web Text))

Max Activating Token Index: 644

Text #19

Max Range: 3.8955. Min Range: -3.8955

Max Act: 3.5826. Min Act: -0.1700

Data Index: 1017371 (C4 (Web Text))

Max Activating Token Index: 575