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

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

Neuron 1371 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.4453. Min Range: -7.4453

Max Act: 7.4453. Min Act: -0.1700

Data Index: 1124507 (C4 (Web Text))

Max Activating Token Index: 632

Text #1

Max Range: 7.4453. Min Range: -7.4453

Max Act: 7.2578. Min Act: -0.1700

Data Index: 1033360 (C4 (Web Text))

Max Activating Token Index: 250

Text #2

Max Range: 7.4453. Min Range: -7.4453

Max Act: 7.1328. Min Act: -0.1700

Data Index: 684413 (C4 (Web Text))

Max Activating Token Index: 554

Text #3

Max Range: 7.4453. Min Range: -7.4453

Max Act: 7.0703. Min Act: -0.1700

Data Index: 1240576 (C4 (Web Text))

Max Activating Token Index: 536

Text #4

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.9766. Min Act: -0.1698

Data Index: 233400 (C4 (Web Text))

Max Activating Token Index: 217

Text #5

Max Range: 7.4453. Min Range: -7.4453

Max Act: 7.0391. Min Act: -0.1700

Data Index: 650402 (C4 (Web Text))

Max Activating Token Index: 450

Text #6

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.8203. Min Act: -0.1700

Data Index: 1275385 (C4 (Web Text))

Max Activating Token Index: 243

Text #7

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.8516. Min Act: -0.1700

Data Index: 683201 (C4 (Web Text))

Max Activating Token Index: 645

Text #8

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.7578. Min Act: -0.1700

Data Index: 693422 (C4 (Web Text))

Max Activating Token Index: 52

Text #9

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.8516. Min Act: -0.1700

Data Index: 809261 (C4 (Web Text))

Max Activating Token Index: 364

Text #10

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.7578. Min Act: -0.1700

Data Index: 538856 (C4 (Web Text))

Max Activating Token Index: 806

Text #11

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.6328. Min Act: -0.1700

Data Index: 234425 (C4 (Web Text))

Max Activating Token Index: 833

Text #12

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.5391. Min Act: -0.1700

Data Index: 1068145 (C4 (Web Text))

Max Activating Token Index: 47

Text #13

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.5391. Min Act: -0.1700

Data Index: 537899 (C4 (Web Text))

Max Activating Token Index: 456

Text #14

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.4766. Min Act: -0.1700

Data Index: 1010488 (C4 (Web Text))

Max Activating Token Index: 40

Text #15

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.4766. Min Act: -0.1698

Data Index: 371332 (C4 (Web Text))

Max Activating Token Index: 270

Text #16

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.5078. Min Act: -0.1700

Data Index: 258846 (C4 (Web Text))

Max Activating Token Index: 39

Text #17

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.4453. Min Act: -0.1700

Data Index: 373369 (C4 (Web Text))

Max Activating Token Index: 599

Text #18

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.4453. Min Act: -0.1700

Data Index: 866200 (C4 (Web Text))

Max Activating Token Index: 353

Text #19

Max Range: 7.4453. Min Range: -7.4453

Max Act: 6.4766. Min Act: -0.1700

Data Index: 1103890 (C4 (Web Text))

Max Activating Token Index: 288