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

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

Neuron 117 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: 5.6732. Min Range: -5.6732

Max Act: 5.5795. Min Act: -0.1700

Data Index: 1075651 (C4 (Web Text))

Max Activating Token Index: 804

Text #1

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.6732. Min Act: -0.1700

Data Index: 253272 (C4 (Web Text))

Max Activating Token Index: 631

Text #2

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.5170. Min Act: -0.1700

Data Index: 294919 (C4 (Web Text))

Max Activating Token Index: 331

Text #3

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.4857. Min Act: -0.1700

Data Index: 55509 (C4 (Web Text))

Max Activating Token Index: 866

Text #4

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.4545. Min Act: -0.1700

Data Index: 498353 (C4 (Web Text))

Max Activating Token Index: 740

Text #5

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.5170. Min Act: -0.1700

Data Index: 1147780 (C4 (Web Text))

Max Activating Token Index: 276

Text #6

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.5170. Min Act: -0.1700

Data Index: 294063 (C4 (Web Text))

Max Activating Token Index: 450

Text #7

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2982. Min Act: -0.1700

Data Index: 1228355 (C4 (Web Text))

Max Activating Token Index: 107

Text #8

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.3295. Min Act: -0.1700

Data Index: 308539 (C4 (Web Text))

Max Activating Token Index: 580

Text #9

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2982. Min Act: -0.1700

Data Index: 754056 (C4 (Web Text))

Max Activating Token Index: 853

Text #10

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2982. Min Act: -0.1700

Data Index: 15097 (C4 (Web Text))

Max Activating Token Index: 612

Text #11

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2357. Min Act: -0.1700

Data Index: 441987 (C4 (Web Text))

Max Activating Token Index: 372

Text #12

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.3295. Min Act: -0.1700

Data Index: 106163 (C4 (Web Text))

Max Activating Token Index: 797

Text #13

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.1420. Min Act: -0.1700

Data Index: 1062100 (C4 (Web Text))

Max Activating Token Index: 601

Text #14

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.3295. Min Act: -0.1700

Data Index: 244842 (C4 (Web Text))

Max Activating Token Index: 208

Text #15

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2357. Min Act: -0.1700

Data Index: 56855 (C4 (Web Text))

Max Activating Token Index: 340

Text #16

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2045. Min Act: -0.1700

Data Index: 1291484 (C4 (Web Text))

Max Activating Token Index: 496

Text #17

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2670. Min Act: -0.1700

Data Index: 1057046 (C4 (Web Text))

Max Activating Token Index: 613

Text #18

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.2045. Min Act: -0.1700

Data Index: 642005 (C4 (Web Text))

Max Activating Token Index: 312

Text #19

Max Range: 5.6732. Min Range: -5.6732

Max Act: 5.1420. Min Act: -0.1700

Data Index: 385266 (C4 (Web Text))

Max Activating Token Index: 91