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

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

Neuron 605 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: 6.5849. Min Range: -6.5849

Max Act: 6.5849. Min Act: -0.1700

Data Index: 291990 (C4 (Web Text))

Max Activating Token Index: 889

Text #1

Max Range: 6.5849. Min Range: -6.5849

Max Act: 6.4599. Min Act: -0.1700

Data Index: 1226163 (C4 (Web Text))

Max Activating Token Index: 692

Text #2

Max Range: 6.5849. Min Range: -6.5849

Max Act: 6.3974. Min Act: -0.1700

Data Index: 1157265 (C4 (Web Text))

Max Activating Token Index: 696

Text #3

Max Range: 6.5849. Min Range: -6.5849

Max Act: 6.0537. Min Act: -0.1700

Data Index: 889482 (C4 (Web Text))

Max Activating Token Index: 351

Text #4

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.9599. Min Act: -0.1700

Data Index: 802226 (C4 (Web Text))

Max Activating Token Index: 308

Text #5

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.9287. Min Act: -0.1700

Data Index: 1561139 (Python Code)

Max Activating Token Index: 931

Text #6

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.8974. Min Act: -0.1700

Data Index: 1463738 (Python Code)

Max Activating Token Index: 948

Text #7

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.8974. Min Act: -0.1700

Data Index: 1539967 (Python Code)

Max Activating Token Index: 62

Text #8

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.8349. Min Act: -0.1700

Data Index: 857615 (C4 (Web Text))

Max Activating Token Index: 193

Text #9

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.8037. Min Act: -0.1700

Data Index: 20807 (C4 (Web Text))

Max Activating Token Index: 244

Text #10

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.8037. Min Act: -0.1362

Data Index: 1586482 (Python Code)

Max Activating Token Index: 956

Text #11

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.8037. Min Act: -0.1700

Data Index: 1156036 (C4 (Web Text))

Max Activating Token Index: 798

Text #12

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.7724. Min Act: -0.1700

Data Index: 534522 (C4 (Web Text))

Max Activating Token Index: 490

Text #13

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.7724. Min Act: -0.1700

Data Index: 921185 (C4 (Web Text))

Max Activating Token Index: 113

Text #14

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.7099. Min Act: -0.1699

Data Index: 436968 (C4 (Web Text))

Max Activating Token Index: 420

Text #15

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.7099. Min Act: -0.1700

Data Index: 1332929 (C4 (Web Text))

Max Activating Token Index: 414

Text #16

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.7412. Min Act: -0.1700

Data Index: 882091 (C4 (Web Text))

Max Activating Token Index: 826

Text #17

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.6474. Min Act: -0.1700

Data Index: 637534 (C4 (Web Text))

Max Activating Token Index: 747

Text #18

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.6474. Min Act: -0.1694

Data Index: 1382549 (Python Code)

Max Activating Token Index: 45

Text #19

Max Range: 6.5849. Min Range: -6.5849

Max Act: 5.6787. Min Act: -0.1700

Data Index: 1618809 (Python Code)

Max Activating Token Index: 11