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

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

Neuron 386 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: 7.7733. Min Range: -7.7733

Max Act: 7.7733. Min Act: -0.1700

Data Index: 944190 (C4 (Web Text))

Max Activating Token Index: 707

Text #1

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.2108. Min Act: -0.1699

Data Index: 743766 (C4 (Web Text))

Max Activating Token Index: 565

Text #2

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.1796. Min Act: -0.1700

Data Index: 1225306 (C4 (Web Text))

Max Activating Token Index: 547

Text #3

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.2421. Min Act: -0.1700

Data Index: 213749 (C4 (Web Text))

Max Activating Token Index: 468

Text #4

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.2108. Min Act: -0.1700

Data Index: 1040927 (C4 (Web Text))

Max Activating Token Index: 773

Text #5

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.1171. Min Act: -0.1700

Data Index: 782168 (C4 (Web Text))

Max Activating Token Index: 535

Text #6

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.0858. Min Act: -0.1700

Data Index: 386726 (C4 (Web Text))

Max Activating Token Index: 736

Text #7

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.1796. Min Act: -0.1700

Data Index: 1322377 (C4 (Web Text))

Max Activating Token Index: 926

Text #8

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.0233. Min Act: -0.1700

Data Index: 552920 (C4 (Web Text))

Max Activating Token Index: 470

Text #9

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.9921. Min Act: -0.1700

Data Index: 906604 (C4 (Web Text))

Max Activating Token Index: 835

Text #10

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.9296. Min Act: -0.1700

Data Index: 786425 (C4 (Web Text))

Max Activating Token Index: 31

Text #11

Max Range: 7.7733. Min Range: -7.7733

Max Act: 7.0233. Min Act: -0.1700

Data Index: 1201838 (C4 (Web Text))

Max Activating Token Index: 125

Text #12

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.8983. Min Act: -0.1700

Data Index: 666345 (C4 (Web Text))

Max Activating Token Index: 479

Text #13

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.8671. Min Act: -0.1700

Data Index: 862266 (C4 (Web Text))

Max Activating Token Index: 28

Text #14

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.9608. Min Act: -0.1700

Data Index: 706705 (C4 (Web Text))

Max Activating Token Index: 244

Text #15

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.8983. Min Act: -0.1700

Data Index: 957289 (C4 (Web Text))

Max Activating Token Index: 234

Text #16

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.8671. Min Act: -0.1700

Data Index: 224551 (C4 (Web Text))

Max Activating Token Index: 391

Text #17

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.8046. Min Act: -0.1700

Data Index: 1302916 (C4 (Web Text))

Max Activating Token Index: 46

Text #18

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.7733. Min Act: -0.1700

Data Index: 413281 (C4 (Web Text))

Max Activating Token Index: 35

Text #19

Max Range: 7.7733. Min Range: -7.7733

Max Act: 6.8046. Min Act: -0.1700

Data Index: 149482 (C4 (Web Text))

Max Activating Token Index: 716