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

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

Neuron 49 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: 3.6364. Min Range: -3.6364

Max Act: 3.6364. Min Act: -0.1700

Data Index: 640296 (C4 (Web Text))

Max Activating Token Index: 288

Text #1

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.6051. Min Act: -0.1700

Data Index: 304834 (C4 (Web Text))

Max Activating Token Index: 888

Text #2

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.5581. Min Act: -0.1700

Data Index: 78385 (C4 (Web Text))

Max Activating Token Index: 289

Text #3

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.5268. Min Act: -0.1700

Data Index: 28121 (C4 (Web Text))

Max Activating Token Index: 509

Text #4

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.5112. Min Act: -0.1700

Data Index: 894102 (C4 (Web Text))

Max Activating Token Index: 941

Text #5

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4641. Min Act: -0.1700

Data Index: 483540 (C4 (Web Text))

Max Activating Token Index: 440

Text #6

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3857. Min Act: -0.1700

Data Index: 869834 (C4 (Web Text))

Max Activating Token Index: 726

Text #7

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4328. Min Act: -0.1700

Data Index: 1126489 (C4 (Web Text))

Max Activating Token Index: 970

Text #8

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4328. Min Act: -0.1700

Data Index: 393285 (C4 (Web Text))

Max Activating Token Index: 322

Text #9

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4485. Min Act: -0.1700

Data Index: 991605 (C4 (Web Text))

Max Activating Token Index: 881

Text #10

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4014. Min Act: -0.1700

Data Index: 130956 (C4 (Web Text))

Max Activating Token Index: 213

Text #11

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4014. Min Act: -0.1700

Data Index: 270493 (C4 (Web Text))

Max Activating Token Index: 651

Text #12

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4014. Min Act: -0.1700

Data Index: 962696 (C4 (Web Text))

Max Activating Token Index: 727

Text #13

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.4014. Min Act: -0.1700

Data Index: 850965 (C4 (Web Text))

Max Activating Token Index: 89

Text #14

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3700. Min Act: -0.1700

Data Index: 477762 (C4 (Web Text))

Max Activating Token Index: 565

Text #15

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3700. Min Act: -0.1700

Data Index: 74511 (C4 (Web Text))

Max Activating Token Index: 468

Text #16

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3544. Min Act: -0.1700

Data Index: 691135 (C4 (Web Text))

Max Activating Token Index: 249

Text #17

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3387. Min Act: -0.1700

Data Index: 1101951 (C4 (Web Text))

Max Activating Token Index: 574

Text #18

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3700. Min Act: -0.1700

Data Index: 206339 (C4 (Web Text))

Max Activating Token Index: 977

Text #19

Max Range: 3.6364. Min Range: -3.6364

Max Act: 3.3700. Min Act: -0.1700

Data Index: 515987 (C4 (Web Text))

Max Activating Token Index: 240