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

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

Neuron 278 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: 1.3361. Min Range: -1.3361

Max Act: 1.3361. Min Act: -0.1700

Data Index: 1548859 (Python Code)

Max Activating Token Index: 924

Text #1

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.3053. Min Act: -0.1700

Data Index: 1486541 (Python Code)

Max Activating Token Index: 593

Text #2

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.2259. Min Act: -0.1699

Data Index: 1564090 (Python Code)

Max Activating Token Index: 220

Text #3

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.2303. Min Act: -0.1699

Data Index: 1469380 (Python Code)

Max Activating Token Index: 283

Text #4

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.1731. Min Act: -0.1699

Data Index: 1523484 (Python Code)

Max Activating Token Index: 479

Text #5

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.1907. Min Act: -0.1698

Data Index: 1442156 (Python Code)

Max Activating Token Index: 696

Text #6

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.1775. Min Act: -0.1696

Data Index: 1598811 (Python Code)

Max Activating Token Index: 881

Text #7

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.0897. Min Act: -0.1700

Data Index: 1387539 (Python Code)

Max Activating Token Index: 154

Text #8

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.9765. Min Act: -0.1698

Data Index: 404562 (C4 (Web Text))

Max Activating Token Index: 902

Text #9

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.9830. Min Act: -0.1700

Data Index: 1393231 (Python Code)

Max Activating Token Index: 621

Text #10

Max Range: 1.3361. Min Range: -1.3361

Max Act: 1.0047. Min Act: -0.1700

Data Index: 1437799 (Python Code)

Max Activating Token Index: 679

Text #11

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.9960. Min Act: -0.1700

Data Index: 1404429 (Python Code)

Max Activating Token Index: 392

Text #12

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.9312. Min Act: -0.1699

Data Index: 1646302 (Python Code)

Max Activating Token Index: 556

Text #13

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8841. Min Act: -0.1693

Data Index: 1476577 (Python Code)

Max Activating Token Index: 296

Text #14

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8798. Min Act: -0.1699

Data Index: 522286 (C4 (Web Text))

Max Activating Token Index: 447

Text #15

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8734. Min Act: -0.1700

Data Index: 1541366 (Python Code)

Max Activating Token Index: 435

Text #16

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8628. Min Act: -0.1697

Data Index: 1550695 (Python Code)

Max Activating Token Index: 770

Text #17

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8416. Min Act: -0.1699

Data Index: 38743 (C4 (Web Text))

Max Activating Token Index: 1003

Text #18

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8068. Min Act: -0.1699

Data Index: 1395163 (Python Code)

Max Activating Token Index: 990

Text #19

Max Range: 1.3361. Min Range: -1.3361

Max Act: 0.8205. Min Act: -0.1700

Data Index: 531680 (C4 (Web Text))

Max Activating Token Index: 538