Model: GELU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1615 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.8909. Min Act: -0.1700

Data Index: 956253 (C4 (Web Text))

Max Activating Token Index: 47

Text #1

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.7347. Min Act: -0.1700

Data Index: 213636 (C4 (Web Text))

Max Activating Token Index: 321

Text #2

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.6722. Min Act: -0.1700

Data Index: 327562 (C4 (Web Text))

Max Activating Token Index: 491

Text #3

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.6097. Min Act: -0.1700

Data Index: 329666 (C4 (Web Text))

Max Activating Token Index: 259

Text #4

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.5472. Min Act: -0.1700

Data Index: 333958 (C4 (Web Text))

Max Activating Token Index: 332

Text #5

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4847. Min Act: -0.1697

Data Index: 1283226 (C4 (Web Text))

Max Activating Token Index: 333

Text #6

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4534. Min Act: -0.1700

Data Index: 987866 (C4 (Web Text))

Max Activating Token Index: 802

Text #7

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4534. Min Act: -0.1700

Data Index: 402868 (C4 (Web Text))

Max Activating Token Index: 145

Text #8

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4222. Min Act: -0.1700

Data Index: 883531 (C4 (Web Text))

Max Activating Token Index: 676

Text #9

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4222. Min Act: -0.1700

Data Index: 369671 (C4 (Web Text))

Max Activating Token Index: 191

Text #10

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3597. Min Act: -0.1700

Data Index: 303860 (C4 (Web Text))

Max Activating Token Index: 31

Text #11

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4222. Min Act: -0.1700

Data Index: 58394 (C4 (Web Text))

Max Activating Token Index: 542

Text #12

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3909. Min Act: -0.1700

Data Index: 1204123 (C4 (Web Text))

Max Activating Token Index: 164

Text #13

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4222. Min Act: -0.1700

Data Index: 715450 (C4 (Web Text))

Max Activating Token Index: 617

Text #14

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3909. Min Act: -0.1700

Data Index: 668467 (C4 (Web Text))

Max Activating Token Index: 309

Text #15

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3284. Min Act: -0.1700

Data Index: 8164 (C4 (Web Text))

Max Activating Token Index: 869

Text #16

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3909. Min Act: -0.1699

Data Index: 309938 (C4 (Web Text))

Max Activating Token Index: 673

Text #17

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.4222. Min Act: -0.1677

Data Index: 503622 (C4 (Web Text))

Max Activating Token Index: 594

Text #18

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3597. Min Act: -0.1700

Data Index: 701526 (C4 (Web Text))

Max Activating Token Index: 727

Text #19

Max Range: 5.8909. Min Range: -5.8909

Max Act: 5.3909. Min Act: -0.1700

Data Index: 1136967 (C4 (Web Text))

Max Activating Token Index: 814