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

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

Neuron 1808 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: 5.3203. Min Range: -5.3203

Max Act: 5.3203. Min Act: -0.1700

Data Index: 1129227 (C4 (Web Text))

Max Activating Token Index: 238

Text #1

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.2578. Min Act: -0.1700

Data Index: 467195 (C4 (Web Text))

Max Activating Token Index: 171

Text #2

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.2266. Min Act: -0.1700

Data Index: 239428 (C4 (Web Text))

Max Activating Token Index: 648

Text #3

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.1016. Min Act: -0.1700

Data Index: 877863 (C4 (Web Text))

Max Activating Token Index: 799

Text #4

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.1641. Min Act: -0.1700

Data Index: 656776 (C4 (Web Text))

Max Activating Token Index: 705

Text #5

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.0078. Min Act: -0.1700

Data Index: 671328 (C4 (Web Text))

Max Activating Token Index: 679

Text #6

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.0078. Min Act: -0.1700

Data Index: 837909 (C4 (Web Text))

Max Activating Token Index: 590

Text #7

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.0391. Min Act: -0.1700

Data Index: 1262290 (C4 (Web Text))

Max Activating Token Index: 869

Text #8

Max Range: 5.3203. Min Range: -5.3203

Max Act: 5.0391. Min Act: -0.1698

Data Index: 1118496 (C4 (Web Text))

Max Activating Token Index: 994

Text #9

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.9766. Min Act: -0.1700

Data Index: 1152636 (C4 (Web Text))

Max Activating Token Index: 27

Text #10

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.8828. Min Act: -0.1700

Data Index: 1222707 (C4 (Web Text))

Max Activating Token Index: 978

Text #11

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.9453. Min Act: -0.1700

Data Index: 644251 (C4 (Web Text))

Max Activating Token Index: 334

Text #12

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.9453. Min Act: -0.1700

Data Index: 132478 (C4 (Web Text))

Max Activating Token Index: 198

Text #13

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.9453. Min Act: -0.1699

Data Index: 373060 (C4 (Web Text))

Max Activating Token Index: 412

Text #14

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.9453. Min Act: -0.1700

Data Index: 683830 (C4 (Web Text))

Max Activating Token Index: 799

Text #15

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.9453. Min Act: -0.1700

Data Index: 557842 (C4 (Web Text))

Max Activating Token Index: 322

Text #16

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.8828. Min Act: -0.1700

Data Index: 45959 (C4 (Web Text))

Max Activating Token Index: 1009

Text #17

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.8516. Min Act: -0.1700

Data Index: 628906 (C4 (Web Text))

Max Activating Token Index: 695

Text #18

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.8516. Min Act: -0.1700

Data Index: 851237 (C4 (Web Text))

Max Activating Token Index: 742

Text #19

Max Range: 5.3203. Min Range: -5.3203

Max Act: 4.8516. Min Act: -0.1700

Data Index: 347766 (C4 (Web Text))

Max Activating Token Index: 700