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

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

Neuron 695 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: 3.3688. Min Range: -3.3688

Max Act: 3.3688. Min Act: -0.1700

Data Index: 601494 (C4 (Web Text))

Max Activating Token Index: 586

Text #1

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.3060. Min Act: -0.1700

Data Index: 697956 (C4 (Web Text))

Max Activating Token Index: 796

Text #2

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.2903. Min Act: -0.1700

Data Index: 25291 (C4 (Web Text))

Max Activating Token Index: 680

Text #3

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.2903. Min Act: -0.1700

Data Index: 1050976 (C4 (Web Text))

Max Activating Token Index: 602

Text #4

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.2588. Min Act: -0.1700

Data Index: 368130 (C4 (Web Text))

Max Activating Token Index: 574

Text #5

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.2903. Min Act: -0.1700

Data Index: 949918 (C4 (Web Text))

Max Activating Token Index: 700

Text #6

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.2274. Min Act: -0.1700

Data Index: 1251092 (C4 (Web Text))

Max Activating Token Index: 546

Text #7

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.2117. Min Act: -0.1700

Data Index: 679627 (C4 (Web Text))

Max Activating Token Index: 940

Text #8

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1802. Min Act: -0.1700

Data Index: 43640 (C4 (Web Text))

Max Activating Token Index: 227

Text #9

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1645. Min Act: -0.1700

Data Index: 660259 (C4 (Web Text))

Max Activating Token Index: 529

Text #10

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1487. Min Act: -0.1700

Data Index: 1221699 (C4 (Web Text))

Max Activating Token Index: 739

Text #11

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1014. Min Act: -0.1700

Data Index: 1081337 (C4 (Web Text))

Max Activating Token Index: 622

Text #12

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1172. Min Act: -0.1700

Data Index: 1248597 (C4 (Web Text))

Max Activating Token Index: 458

Text #13

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1172. Min Act: -0.1700

Data Index: 868256 (C4 (Web Text))

Max Activating Token Index: 856

Text #14

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1014. Min Act: -0.1700

Data Index: 506586 (C4 (Web Text))

Max Activating Token Index: 1003

Text #15

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1487. Min Act: -0.1700

Data Index: 1175835 (C4 (Web Text))

Max Activating Token Index: 451

Text #16

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.1014. Min Act: -0.1700

Data Index: 34479 (C4 (Web Text))

Max Activating Token Index: 752

Text #17

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.0541. Min Act: -0.1700

Data Index: 330364 (C4 (Web Text))

Max Activating Token Index: 925

Text #18

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.0699. Min Act: -0.1700

Data Index: 352450 (C4 (Web Text))

Max Activating Token Index: 775

Text #19

Max Range: 3.3688. Min Range: -3.3688

Max Act: 3.0699. Min Act: -0.1700

Data Index: 629376 (C4 (Web Text))

Max Activating Token Index: 447