Model: GELU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 444 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.4814. Min Act: -0.1700

Data Index: 1415622 (Python Code)

Max Activating Token Index: 596

Text #1

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.4658. Min Act: -0.1700

Data Index: 1589330 (Python Code)

Max Activating Token Index: 433

Text #2

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.4344. Min Act: -0.1700

Data Index: 1646862 (Python Code)

Max Activating Token Index: 959

Text #3

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.4031. Min Act: -0.1700

Data Index: 1445669 (Python Code)

Max Activating Token Index: 397

Text #4

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.4187. Min Act: -0.1700

Data Index: 1487961 (Python Code)

Max Activating Token Index: 788

Text #5

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.4031. Min Act: -0.1700

Data Index: 1616530 (Python Code)

Max Activating Token Index: 562

Text #6

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 1459762 (Python Code)

Max Activating Token Index: 611

Text #7

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 1480224 (Python Code)

Max Activating Token Index: 234

Text #8

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 476485 (C4 (Web Text))

Max Activating Token Index: 164

Text #9

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.2932. Min Act: -0.1700

Data Index: 1362169 (Python Code)

Max Activating Token Index: 854

Text #10

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 170387 (C4 (Web Text))

Max Activating Token Index: 638

Text #11

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 1644774 (Python Code)

Max Activating Token Index: 298

Text #12

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 1479342 (Python Code)

Max Activating Token Index: 722

Text #13

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 1382081 (Python Code)

Max Activating Token Index: 479

Text #14

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3246. Min Act: -0.1700

Data Index: 1384830 (Python Code)

Max Activating Token Index: 821

Text #15

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3089. Min Act: -0.1700

Data Index: 1467850 (Python Code)

Max Activating Token Index: 670

Text #16

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3403. Min Act: -0.1700

Data Index: 1558321 (Python Code)

Max Activating Token Index: 782

Text #17

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3246. Min Act: -0.1700

Data Index: 936903 (C4 (Web Text))

Max Activating Token Index: 503

Text #18

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3089. Min Act: -0.1700

Data Index: 237663 (C4 (Web Text))

Max Activating Token Index: 657

Text #19

Max Range: 3.4814. Min Range: -3.4814

Max Act: 3.3246. Min Act: -0.1700

Data Index: 106776 (C4 (Web Text))

Max Activating Token Index: 643