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

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

Neuron 91 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.5237. Min Act: -0.1700

Data Index: 1291464 (C4 (Web Text))

Max Activating Token Index: 703

Text #1

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.3669. Min Act: -0.1700

Data Index: 614436 (C4 (Web Text))

Max Activating Token Index: 621

Text #2

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.3198. Min Act: -0.1700

Data Index: 870948 (C4 (Web Text))

Max Activating Token Index: 531

Text #3

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2884. Min Act: -0.1700

Data Index: 722102 (C4 (Web Text))

Max Activating Token Index: 694

Text #4

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2727. Min Act: -0.1700

Data Index: 875404 (C4 (Web Text))

Max Activating Token Index: 150

Text #5

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2727. Min Act: -0.1700

Data Index: 5139 (C4 (Web Text))

Max Activating Token Index: 798

Text #6

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2255. Min Act: -0.1700

Data Index: 671070 (C4 (Web Text))

Max Activating Token Index: 510

Text #7

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2570. Min Act: -0.1700

Data Index: 490400 (C4 (Web Text))

Max Activating Token Index: 197

Text #8

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2570. Min Act: -0.1700

Data Index: 383571 (C4 (Web Text))

Max Activating Token Index: 774

Text #9

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1941. Min Act: -0.1700

Data Index: 1495509 (Python Code)

Max Activating Token Index: 845

Text #10

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2412. Min Act: -0.1700

Data Index: 205614 (C4 (Web Text))

Max Activating Token Index: 317

Text #11

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1941. Min Act: -0.1700

Data Index: 800815 (C4 (Web Text))

Max Activating Token Index: 576

Text #12

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1941. Min Act: -0.1700

Data Index: 1343441 (C4 (Web Text))

Max Activating Token Index: 905

Text #13

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2098. Min Act: -0.1700

Data Index: 1031595 (C4 (Web Text))

Max Activating Token Index: 989

Text #14

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1783. Min Act: -0.1700

Data Index: 4236 (C4 (Web Text))

Max Activating Token Index: 38

Text #15

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1626. Min Act: -0.1700

Data Index: 1257298 (C4 (Web Text))

Max Activating Token Index: 861

Text #16

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1468. Min Act: -0.1700

Data Index: 450093 (C4 (Web Text))

Max Activating Token Index: 806

Text #17

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.2098. Min Act: -0.1700

Data Index: 1031638 (C4 (Web Text))

Max Activating Token Index: 709

Text #18

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1941. Min Act: -0.1700

Data Index: 1316025 (C4 (Web Text))

Max Activating Token Index: 613

Text #19

Max Range: 3.5237. Min Range: -3.5237

Max Act: 3.1783. Min Act: -0.1700

Data Index: 366744 (C4 (Web Text))

Max Activating Token Index: 80