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

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

Neuron 1641 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-4l')

Text #0

Max Range: 4.1066. Min Range: -4.1066

Max Act: 4.1066. Min Act: -0.1700

Data Index: 635764 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 4.1066. Min Range: -4.1066

Max Act: 4.0441. Min Act: -0.1700

Data Index: 1059165 (C4 (Web Text))

Max Activating Token Index: 172

Text #2

Max Range: 4.1066. Min Range: -4.1066

Max Act: 4.0129. Min Act: -0.1700

Data Index: 935143 (C4 (Web Text))

Max Activating Token Index: 247

Text #3

Max Range: 4.1066. Min Range: -4.1066

Max Act: 4.0129. Min Act: -0.1700

Data Index: 213955 (C4 (Web Text))

Max Activating Token Index: 1015

Text #4

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 916713 (C4 (Web Text))

Max Activating Token Index: 821

Text #5

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 1296129 (C4 (Web Text))

Max Activating Token Index: 301

Text #6

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9816. Min Act: -0.1700

Data Index: 824972 (C4 (Web Text))

Max Activating Token Index: 326

Text #7

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 1294963 (C4 (Web Text))

Max Activating Token Index: 667

Text #8

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 1282751 (C4 (Web Text))

Max Activating Token Index: 534

Text #9

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 225599 (C4 (Web Text))

Max Activating Token Index: 592

Text #10

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 1487997 (Python Code)

Max Activating Token Index: 5

Text #11

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 1207517 (C4 (Web Text))

Max Activating Token Index: 590

Text #12

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9503. Min Act: -0.1700

Data Index: 1076755 (C4 (Web Text))

Max Activating Token Index: 457

Text #13

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9191. Min Act: -0.1700

Data Index: 392107 (C4 (Web Text))

Max Activating Token Index: 7

Text #14

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9191. Min Act: -0.1700

Data Index: 835520 (C4 (Web Text))

Max Activating Token Index: 681

Text #15

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9191. Min Act: -0.1700

Data Index: 971442 (C4 (Web Text))

Max Activating Token Index: 833

Text #16

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.9191. Min Act: -0.1700

Data Index: 547306 (C4 (Web Text))

Max Activating Token Index: 199

Text #17

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.8878. Min Act: -0.1700

Data Index: 837167 (C4 (Web Text))

Max Activating Token Index: 408

Text #18

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.8878. Min Act: -0.1700

Data Index: 573665 (C4 (Web Text))

Max Activating Token Index: 693

Text #19

Max Range: 4.1066. Min Range: -4.1066

Max Act: 3.8878. Min Act: -0.1700

Data Index: 189574 (C4 (Web Text))

Max Activating Token Index: 176