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

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

Neuron 1072 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: 5.1485. Min Range: -5.1485

Max Act: 5.1485. Min Act: -0.1700

Data Index: 657617 (C4 (Web Text))

Max Activating Token Index: 577

Text #1

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.8985. Min Act: -0.1700

Data Index: 1245678 (C4 (Web Text))

Max Activating Token Index: 527

Text #2

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.8360. Min Act: -0.1700

Data Index: 201369 (C4 (Web Text))

Max Activating Token Index: 607

Text #3

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.6172. Min Act: -0.1700

Data Index: 956818 (C4 (Web Text))

Max Activating Token Index: 928

Text #4

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.5860. Min Act: -0.1700

Data Index: 1104201 (C4 (Web Text))

Max Activating Token Index: 20

Text #5

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.5860. Min Act: -0.1700

Data Index: 846490 (C4 (Web Text))

Max Activating Token Index: 308

Text #6

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.5235. Min Act: -0.1700

Data Index: 426493 (C4 (Web Text))

Max Activating Token Index: 319

Text #7

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.5235. Min Act: -0.1700

Data Index: 610750 (C4 (Web Text))

Max Activating Token Index: 547

Text #8

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.4297. Min Act: -0.1700

Data Index: 383894 (C4 (Web Text))

Max Activating Token Index: 112

Text #9

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.4609. Min Act: -0.1700

Data Index: 160175 (C4 (Web Text))

Max Activating Token Index: 620

Text #10

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.4297. Min Act: -0.1700

Data Index: 664325 (C4 (Web Text))

Max Activating Token Index: 77

Text #11

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.4609. Min Act: -0.1700

Data Index: 940352 (C4 (Web Text))

Max Activating Token Index: 528

Text #12

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.4609. Min Act: -0.1700

Data Index: 823800 (C4 (Web Text))

Max Activating Token Index: 757

Text #13

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.3984. Min Act: -0.1700

Data Index: 899880 (C4 (Web Text))

Max Activating Token Index: 32

Text #14

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.4297. Min Act: -0.1700

Data Index: 453542 (C4 (Web Text))

Max Activating Token Index: 529

Text #15

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.3359. Min Act: -0.1700

Data Index: 513755 (C4 (Web Text))

Max Activating Token Index: 873

Text #16

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.3672. Min Act: -0.1700

Data Index: 77629 (C4 (Web Text))

Max Activating Token Index: 695

Text #17

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.3984. Min Act: -0.1700

Data Index: 966514 (C4 (Web Text))

Max Activating Token Index: 420

Text #18

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.3672. Min Act: -0.1700

Data Index: 265980 (C4 (Web Text))

Max Activating Token Index: 12

Text #19

Max Range: 5.1485. Min Range: -5.1485

Max Act: 4.3672. Min Act: -0.1700

Data Index: 952705 (C4 (Web Text))

Max Activating Token Index: 141