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

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

Neuron 40 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: 3.7755. Min Range: -3.7755

Max Act: 3.7755. Min Act: -0.1700

Data Index: 1128435 (C4 (Web Text))

Max Activating Token Index: 275

Text #1

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.7129. Min Act: -0.1700

Data Index: 814660 (C4 (Web Text))

Max Activating Token Index: 666

Text #2

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6816. Min Act: -0.1700

Data Index: 1112654 (C4 (Web Text))

Max Activating Token Index: 219

Text #3

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6816. Min Act: -0.1700

Data Index: 1025014 (C4 (Web Text))

Max Activating Token Index: 190

Text #4

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6503. Min Act: -0.1700

Data Index: 763516 (C4 (Web Text))

Max Activating Token Index: 559

Text #5

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5877. Min Act: -0.1700

Data Index: 13788 (C4 (Web Text))

Max Activating Token Index: 395

Text #6

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6190. Min Act: -0.1700

Data Index: 445202 (C4 (Web Text))

Max Activating Token Index: 600

Text #7

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6190. Min Act: -0.1700

Data Index: 870552 (C4 (Web Text))

Max Activating Token Index: 406

Text #8

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6190. Min Act: -0.1700

Data Index: 1000405 (C4 (Web Text))

Max Activating Token Index: 386

Text #9

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5877. Min Act: -0.1700

Data Index: 597331 (C4 (Web Text))

Max Activating Token Index: 547

Text #10

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 170430 (C4 (Web Text))

Max Activating Token Index: 281

Text #11

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.6503. Min Act: -0.1700

Data Index: 62236 (C4 (Web Text))

Max Activating Token Index: 901

Text #12

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5877. Min Act: -0.1700

Data Index: 442308 (C4 (Web Text))

Max Activating Token Index: 362

Text #13

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5877. Min Act: -0.1700

Data Index: 505092 (C4 (Web Text))

Max Activating Token Index: 432

Text #14

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 100161 (C4 (Web Text))

Max Activating Token Index: 233

Text #15

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 124219 (C4 (Web Text))

Max Activating Token Index: 182

Text #16

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 999972 (C4 (Web Text))

Max Activating Token Index: 910

Text #17

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 1021644 (C4 (Web Text))

Max Activating Token Index: 527

Text #18

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 293764 (C4 (Web Text))

Max Activating Token Index: 601

Text #19

Max Range: 3.7755. Min Range: -3.7755

Max Act: 3.5564. Min Act: -0.1700

Data Index: 754584 (C4 (Web Text))

Max Activating Token Index: 623