Model: GELU Model: 2 Layers, 2048 Neurons per Layer

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

Neuron 507 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-2l')

Text #0

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.9107. Min Act: -0.1700

Data Index: 1636809 (Python Code)

Max Activating Token Index: 120

Text #1

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.8169. Min Act: -0.1700

Data Index: 1548646 (Python Code)

Max Activating Token Index: 79

Text #2

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.6917. Min Act: -0.1700

Data Index: 855280 (C4 (Web Text))

Max Activating Token Index: 809

Text #3

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.7230. Min Act: -0.1700

Data Index: 1419218 (Python Code)

Max Activating Token Index: 269

Text #4

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.6917. Min Act: -0.1700

Data Index: 1185122 (C4 (Web Text))

Max Activating Token Index: 718

Text #5

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.6448. Min Act: -0.1700

Data Index: 1592340 (Python Code)

Max Activating Token Index: 200

Text #6

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.6291. Min Act: -0.1700

Data Index: 948615 (C4 (Web Text))

Max Activating Token Index: 303

Text #7

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.6135. Min Act: -0.1700

Data Index: 1564127 (Python Code)

Max Activating Token Index: 254

Text #8

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.5822. Min Act: -0.1700

Data Index: 466768 (C4 (Web Text))

Max Activating Token Index: 430

Text #9

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.5665. Min Act: -0.1700

Data Index: 323258 (C4 (Web Text))

Max Activating Token Index: 585

Text #10

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.5195. Min Act: -0.1700

Data Index: 363950 (C4 (Web Text))

Max Activating Token Index: 208

Text #11

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.5195. Min Act: -0.1700

Data Index: 1412083 (Python Code)

Max Activating Token Index: 782

Text #12

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.5195. Min Act: -0.1700

Data Index: 1536704 (Python Code)

Max Activating Token Index: 685

Text #13

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4725. Min Act: -0.1700

Data Index: 54587 (C4 (Web Text))

Max Activating Token Index: 905

Text #14

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4725. Min Act: -0.1700

Data Index: 555030 (C4 (Web Text))

Max Activating Token Index: 658

Text #15

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4725. Min Act: -0.1700

Data Index: 281436 (C4 (Web Text))

Max Activating Token Index: 701

Text #16

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4568. Min Act: -0.1700

Data Index: 1580946 (Python Code)

Max Activating Token Index: 198

Text #17

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4568. Min Act: -0.1696

Data Index: 1537073 (Python Code)

Max Activating Token Index: 567

Text #18

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4412. Min Act: -0.1700

Data Index: 1498759 (Python Code)

Max Activating Token Index: 710

Text #19

Max Range: 3.9107. Min Range: -3.9107

Max Act: 3.4255. Min Act: -0.1700

Data Index: 1175559 (C4 (Web Text))

Max Activating Token Index: 978