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

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

Neuron 1887 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8714. Min Act: -0.1700

Data Index: 1253980 (C4 (Web Text))

Max Activating Token Index: 935

Text #1

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8402. Min Act: -0.1700

Data Index: 839458 (C4 (Web Text))

Max Activating Token Index: 286

Text #2

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8245. Min Act: -0.1700

Data Index: 617887 (C4 (Web Text))

Max Activating Token Index: 10

Text #3

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8245. Min Act: -0.1700

Data Index: 805645 (C4 (Web Text))

Max Activating Token Index: 862

Text #4

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8245. Min Act: -0.1700

Data Index: 46619 (C4 (Web Text))

Max Activating Token Index: 124

Text #5

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7932. Min Act: -0.1700

Data Index: 200749 (C4 (Web Text))

Max Activating Token Index: 152

Text #6

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8245. Min Act: -0.1700

Data Index: 1097291 (C4 (Web Text))

Max Activating Token Index: 609

Text #7

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8245. Min Act: -0.1700

Data Index: 785365 (C4 (Web Text))

Max Activating Token Index: 1022

Text #8

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8089. Min Act: -0.1700

Data Index: 273921 (C4 (Web Text))

Max Activating Token Index: 773

Text #9

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8089. Min Act: -0.1700

Data Index: 785771 (C4 (Web Text))

Max Activating Token Index: 157

Text #10

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7932. Min Act: -0.1700

Data Index: 826029 (C4 (Web Text))

Max Activating Token Index: 76

Text #11

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.8089. Min Act: -0.1700

Data Index: 229132 (C4 (Web Text))

Max Activating Token Index: 3

Text #12

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7932. Min Act: -0.1700

Data Index: 1414999 (Python Code)

Max Activating Token Index: 56

Text #13

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7932. Min Act: -0.1700

Data Index: 990372 (C4 (Web Text))

Max Activating Token Index: 709

Text #14

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7776. Min Act: -0.1700

Data Index: 653324 (C4 (Web Text))

Max Activating Token Index: 70

Text #15

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7776. Min Act: -0.1700

Data Index: 543633 (C4 (Web Text))

Max Activating Token Index: 263

Text #16

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7463. Min Act: -0.1700

Data Index: 805867 (C4 (Web Text))

Max Activating Token Index: 965

Text #17

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7776. Min Act: -0.1700

Data Index: 667740 (C4 (Web Text))

Max Activating Token Index: 432

Text #18

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7463. Min Act: -0.1700

Data Index: 430772 (C4 (Web Text))

Max Activating Token Index: 753

Text #19

Max Range: 3.8714. Min Range: -3.8714

Max Act: 3.7619. Min Act: -0.1700

Data Index: 996486 (C4 (Web Text))

Max Activating Token Index: 871