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

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

Neuron 224 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.4695. Min Act: -0.1700

Data Index: 957168 (C4 (Web Text))

Max Activating Token Index: 440

Text #1

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.4383. Min Act: -0.1700

Data Index: 375121 (C4 (Web Text))

Max Activating Token Index: 343

Text #2

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.4383. Min Act: -0.1700

Data Index: 197857 (C4 (Web Text))

Max Activating Token Index: 213

Text #3

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.3758. Min Act: -0.1699

Data Index: 471095 (C4 (Web Text))

Max Activating Token Index: 438

Text #4

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.2820. Min Act: -0.1700

Data Index: 542993 (C4 (Web Text))

Max Activating Token Index: 387

Text #5

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.3133. Min Act: -0.1700

Data Index: 932961 (C4 (Web Text))

Max Activating Token Index: 723

Text #6

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.2820. Min Act: -0.1700

Data Index: 352332 (C4 (Web Text))

Max Activating Token Index: 655

Text #7

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.2820. Min Act: -0.1700

Data Index: 1003533 (C4 (Web Text))

Max Activating Token Index: 427

Text #8

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.1570. Min Act: -0.1699

Data Index: 232800 (C4 (Web Text))

Max Activating Token Index: 606

Text #9

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0945. Min Act: -0.1700

Data Index: 570668 (C4 (Web Text))

Max Activating Token Index: 975

Text #10

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.1258. Min Act: -0.1700

Data Index: 1313958 (C4 (Web Text))

Max Activating Token Index: 606

Text #11

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.1570. Min Act: -0.1700

Data Index: 284239 (C4 (Web Text))

Max Activating Token Index: 930

Text #12

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.1570. Min Act: -0.1700

Data Index: 485996 (C4 (Web Text))

Max Activating Token Index: 379

Text #13

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.1258. Min Act: -0.1700

Data Index: 163251 (C4 (Web Text))

Max Activating Token Index: 450

Text #14

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0945. Min Act: -0.1700

Data Index: 244037 (C4 (Web Text))

Max Activating Token Index: 282

Text #15

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0945. Min Act: -0.1700

Data Index: 419064 (C4 (Web Text))

Max Activating Token Index: 791

Text #16

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0320. Min Act: -0.1700

Data Index: 341611 (C4 (Web Text))

Max Activating Token Index: 968

Text #17

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0633. Min Act: -0.1700

Data Index: 653328 (C4 (Web Text))

Max Activating Token Index: 702

Text #18

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0633. Min Act: -0.1700

Data Index: 419509 (C4 (Web Text))

Max Activating Token Index: 875

Text #19

Max Range: 8.4695. Min Range: -8.4695

Max Act: 8.0320. Min Act: -0.1700

Data Index: 1447760 (Python Code)

Max Activating Token Index: 347