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

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

Neuron 1401 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.1403. Min Range: -6.1403

Max Act: 6.1403. Min Act: -0.1700

Data Index: 258518 (C4 (Web Text))

Max Activating Token Index: 524

Text #1

Max Range: 6.1403. Min Range: -6.1403

Max Act: 6.1090. Min Act: -0.1700

Data Index: 435161 (C4 (Web Text))

Max Activating Token Index: 144

Text #2

Max Range: 6.1403. Min Range: -6.1403

Max Act: 6.0153. Min Act: -0.1700

Data Index: 634631 (C4 (Web Text))

Max Activating Token Index: 250

Text #3

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.9840. Min Act: -0.1699

Data Index: 563391 (C4 (Web Text))

Max Activating Token Index: 1019

Text #4

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.9528. Min Act: -0.1700

Data Index: 656228 (C4 (Web Text))

Max Activating Token Index: 416

Text #5

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.7965. Min Act: -0.1700

Data Index: 1067698 (C4 (Web Text))

Max Activating Token Index: 1006

Text #6

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6715. Min Act: -0.1700

Data Index: 220228 (C4 (Web Text))

Max Activating Token Index: 669

Text #7

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.7028. Min Act: -0.1700

Data Index: 481123 (C4 (Web Text))

Max Activating Token Index: 975

Text #8

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.7028. Min Act: -0.1700

Data Index: 1180741 (C4 (Web Text))

Max Activating Token Index: 999

Text #9

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6090. Min Act: -0.1700

Data Index: 779535 (C4 (Web Text))

Max Activating Token Index: 750

Text #10

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6403. Min Act: -0.1700

Data Index: 77419 (C4 (Web Text))

Max Activating Token Index: 855

Text #11

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6403. Min Act: -0.1700

Data Index: 542254 (C4 (Web Text))

Max Activating Token Index: 748

Text #12

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6715. Min Act: -0.1700

Data Index: 52211 (C4 (Web Text))

Max Activating Token Index: 306

Text #13

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6403. Min Act: -0.1700

Data Index: 571455 (C4 (Web Text))

Max Activating Token Index: 821

Text #14

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.5465. Min Act: -0.1700

Data Index: 560154 (C4 (Web Text))

Max Activating Token Index: 510

Text #15

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.5778. Min Act: -0.1700

Data Index: 911930 (C4 (Web Text))

Max Activating Token Index: 201

Text #16

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.5465. Min Act: -0.1700

Data Index: 402314 (C4 (Web Text))

Max Activating Token Index: 910

Text #17

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.5778. Min Act: -0.1700

Data Index: 370904 (C4 (Web Text))

Max Activating Token Index: 234

Text #18

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6090. Min Act: -0.1700

Data Index: 214022 (C4 (Web Text))

Max Activating Token Index: 528

Text #19

Max Range: 6.1403. Min Range: -6.1403

Max Act: 5.6090. Min Act: -0.1700

Data Index: 422259 (C4 (Web Text))

Max Activating Token Index: 660