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

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

Neuron 189 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: 3.4362. Min Range: -3.4362

Max Act: 3.4362. Min Act: -0.1700

Data Index: 1417896 (Python Code)

Max Activating Token Index: 164

Text #1

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.2164. Min Act: -0.1700

Data Index: 1329215 (C4 (Web Text))

Max Activating Token Index: 813

Text #2

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1534. Min Act: -0.1700

Data Index: 1554100 (Python Code)

Max Activating Token Index: 735

Text #3

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1849. Min Act: -0.1700

Data Index: 1476498 (Python Code)

Max Activating Token Index: 31

Text #4

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1691. Min Act: -0.1700

Data Index: 270132 (C4 (Web Text))

Max Activating Token Index: 45

Text #5

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1691. Min Act: -0.1700

Data Index: 1600638 (Python Code)

Max Activating Token Index: 967

Text #6

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1534. Min Act: -0.1700

Data Index: 940465 (C4 (Web Text))

Max Activating Token Index: 418

Text #7

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1377. Min Act: -0.1700

Data Index: 1364858 (Python Code)

Max Activating Token Index: 684

Text #8

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1534. Min Act: -0.1700

Data Index: 117827 (C4 (Web Text))

Max Activating Token Index: 455

Text #9

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1377. Min Act: -0.1700

Data Index: 584409 (C4 (Web Text))

Max Activating Token Index: 633

Text #10

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1377. Min Act: -0.1700

Data Index: 1627010 (Python Code)

Max Activating Token Index: 361

Text #11

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1219. Min Act: -0.1700

Data Index: 1381568 (Python Code)

Max Activating Token Index: 826

Text #12

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1219. Min Act: -0.1700

Data Index: 1551158 (Python Code)

Max Activating Token Index: 699

Text #13

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1061. Min Act: -0.1700

Data Index: 1411747 (Python Code)

Max Activating Token Index: 7

Text #14

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1061. Min Act: -0.1700

Data Index: 83802 (C4 (Web Text))

Max Activating Token Index: 251

Text #15

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1061. Min Act: -0.1700

Data Index: 1459708 (Python Code)

Max Activating Token Index: 265

Text #16

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1061. Min Act: -0.1700

Data Index: 874287 (C4 (Web Text))

Max Activating Token Index: 416

Text #17

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.0904. Min Act: -0.1700

Data Index: 1146013 (C4 (Web Text))

Max Activating Token Index: 540

Text #18

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.1061. Min Act: -0.1700

Data Index: 1484765 (Python Code)

Max Activating Token Index: 890

Text #19

Max Range: 3.4362. Min Range: -3.4362

Max Act: 3.0746. Min Act: -0.1700

Data Index: 1490653 (Python Code)

Max Activating Token Index: 825