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

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

Neuron 2013 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.7151. Min Act: -0.1700

Data Index: 68643 (C4 (Web Text))

Max Activating Token Index: 723

Text #1

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.7620. Min Act: -0.1700

Data Index: 1138918 (C4 (Web Text))

Max Activating Token Index: 947

Text #2

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.7464. Min Act: -0.1700

Data Index: 998702 (C4 (Web Text))

Max Activating Token Index: 46

Text #3

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.7151. Min Act: -0.1700

Data Index: 456712 (C4 (Web Text))

Max Activating Token Index: 118

Text #4

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.7307. Min Act: -0.1700

Data Index: 846233 (C4 (Web Text))

Max Activating Token Index: 545

Text #5

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6994. Min Act: -0.1700

Data Index: 598981 (C4 (Web Text))

Max Activating Token Index: 200

Text #6

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6994. Min Act: -0.1700

Data Index: 971681 (C4 (Web Text))

Max Activating Token Index: 303

Text #7

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6994. Min Act: -0.1700

Data Index: 482862 (C4 (Web Text))

Max Activating Token Index: 201

Text #8

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6838. Min Act: -0.1700

Data Index: 1018406 (C4 (Web Text))

Max Activating Token Index: 246

Text #9

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6838. Min Act: -0.1700

Data Index: 1342670 (C4 (Web Text))

Max Activating Token Index: 674

Text #10

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6525. Min Act: -0.1700

Data Index: 806938 (C4 (Web Text))

Max Activating Token Index: 481

Text #11

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6525. Min Act: -0.1700

Data Index: 785891 (C4 (Web Text))

Max Activating Token Index: 324

Text #12

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6525. Min Act: -0.1700

Data Index: 1284027 (C4 (Web Text))

Max Activating Token Index: 46

Text #13

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6368. Min Act: -0.1700

Data Index: 157832 (C4 (Web Text))

Max Activating Token Index: 908

Text #14

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.5898. Min Act: -0.1700

Data Index: 491922 (C4 (Web Text))

Max Activating Token Index: 770

Text #15

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.5429. Min Act: -0.1700

Data Index: 376468 (C4 (Web Text))

Max Activating Token Index: 730

Text #16

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.5742. Min Act: -0.1700

Data Index: 527994 (C4 (Web Text))

Max Activating Token Index: 177

Text #17

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.5898. Min Act: -0.1700

Data Index: 477521 (C4 (Web Text))

Max Activating Token Index: 382

Text #18

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.6055. Min Act: -0.1700

Data Index: 1244636 (C4 (Web Text))

Max Activating Token Index: 833

Text #19

Max Range: 3.7620. Min Range: -3.7620

Max Act: 3.5585. Min Act: -0.1700

Data Index: 1131152 (C4 (Web Text))

Max Activating Token Index: 968