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

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

Neuron 909 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: 3.5464. Min Range: -3.5464

Max Act: 3.5464. Min Act: -0.1700

Data Index: 1606877 (Python Code)

Max Activating Token Index: 566

Text #1

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3740. Min Act: -0.1700

Data Index: 1591497 (Python Code)

Max Activating Token Index: 391

Text #2

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3426. Min Act: -0.1700

Data Index: 1499944 (Python Code)

Max Activating Token Index: 193

Text #3

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3269. Min Act: -0.1700

Data Index: 1258094 (C4 (Web Text))

Max Activating Token Index: 754

Text #4

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3269. Min Act: -0.1700

Data Index: 1326047 (C4 (Web Text))

Max Activating Token Index: 703

Text #5

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 1321448 (C4 (Web Text))

Max Activating Token Index: 716

Text #6

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 1581801 (Python Code)

Max Activating Token Index: 319

Text #7

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 1293821 (C4 (Web Text))

Max Activating Token Index: 748

Text #8

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 628155 (C4 (Web Text))

Max Activating Token Index: 855

Text #9

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2955. Min Act: -0.1700

Data Index: 561377 (C4 (Web Text))

Max Activating Token Index: 754

Text #10

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 1173478 (C4 (Web Text))

Max Activating Token Index: 537

Text #11

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 776226 (C4 (Web Text))

Max Activating Token Index: 889

Text #12

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 439029 (C4 (Web Text))

Max Activating Token Index: 463

Text #13

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.3112. Min Act: -0.1700

Data Index: 796587 (C4 (Web Text))

Max Activating Token Index: 765

Text #14

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2955. Min Act: -0.1700

Data Index: 1426863 (Python Code)

Max Activating Token Index: 539

Text #15

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2955. Min Act: -0.1700

Data Index: 1192425 (C4 (Web Text))

Max Activating Token Index: 640

Text #16

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2798. Min Act: -0.1700

Data Index: 969923 (C4 (Web Text))

Max Activating Token Index: 433

Text #17

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2798. Min Act: -0.1700

Data Index: 140443 (C4 (Web Text))

Max Activating Token Index: 269

Text #18

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2798. Min Act: -0.1700

Data Index: 993913 (C4 (Web Text))

Max Activating Token Index: 711

Text #19

Max Range: 3.5464. Min Range: -3.5464

Max Act: 3.2798. Min Act: -0.1700

Data Index: 135633 (C4 (Web Text))

Max Activating Token Index: 645