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

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

Neuron 701 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.1048. Min Range: -5.1048

Max Act: 5.1048. Min Act: -0.1699

Data Index: 1639110 (Python Code)

Max Activating Token Index: 773

Text #1

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.9798. Min Act: -0.1699

Data Index: 1458589 (Python Code)

Max Activating Token Index: 652

Text #2

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.9134. Min Act: -0.1699

Data Index: 1574224 (Python Code)

Max Activating Token Index: 813

Text #3

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.8118. Min Act: -0.1699

Data Index: 1500887 (Python Code)

Max Activating Token Index: 456

Text #4

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.7923. Min Act: -0.1699

Data Index: 1508339 (Python Code)

Max Activating Token Index: 666

Text #5

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.6790. Min Act: -0.1699

Data Index: 1611346 (Python Code)

Max Activating Token Index: 856

Text #6

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.6126. Min Act: -0.1699

Data Index: 1454648 (Python Code)

Max Activating Token Index: 432

Text #7

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.6360. Min Act: -0.1699

Data Index: 1617246 (Python Code)

Max Activating Token Index: 565

Text #8

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5969. Min Act: -0.1699

Data Index: 1523142 (Python Code)

Max Activating Token Index: 552

Text #9

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5891. Min Act: -0.1699

Data Index: 1499140 (Python Code)

Max Activating Token Index: 302

Text #10

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5735. Min Act: -0.1699

Data Index: 1439904 (Python Code)

Max Activating Token Index: 884

Text #11

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5462. Min Act: -0.1699

Data Index: 1614667 (Python Code)

Max Activating Token Index: 298

Text #12

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5462. Min Act: -0.1699

Data Index: 1463675 (Python Code)

Max Activating Token Index: 1001

Text #13

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5696. Min Act: -0.1699

Data Index: 1516149 (Python Code)

Max Activating Token Index: 778

Text #14

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5110. Min Act: -0.1699

Data Index: 1587328 (Python Code)

Max Activating Token Index: 708

Text #15

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.5032. Min Act: -0.1699

Data Index: 1512976 (Python Code)

Max Activating Token Index: 478

Text #16

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.4797. Min Act: -0.1699

Data Index: 1582747 (Python Code)

Max Activating Token Index: 616

Text #17

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.4563. Min Act: -0.1699

Data Index: 1369011 (Python Code)

Max Activating Token Index: 910

Text #18

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.4719. Min Act: -0.1699

Data Index: 1523162 (Python Code)

Max Activating Token Index: 518

Text #19

Max Range: 5.1048. Min Range: -5.1048

Max Act: 4.4563. Min Act: -0.1699

Data Index: 1460812 (Python Code)

Max Activating Token Index: 613