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

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

Neuron 2027 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: 3.5698. Min Range: -3.5698

Max Act: 3.5698. Min Act: -0.1700

Data Index: 1538383 (Python Code)

Max Activating Token Index: 5

Text #1

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.5385. Min Act: -0.1700

Data Index: 346827 (C4 (Web Text))

Max Activating Token Index: 84

Text #2

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.5228. Min Act: -0.1700

Data Index: 1550445 (Python Code)

Max Activating Token Index: 5

Text #3

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.5072. Min Act: -0.1700

Data Index: 1440195 (Python Code)

Max Activating Token Index: 354

Text #4

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.5072. Min Act: -0.1700

Data Index: 1477074 (Python Code)

Max Activating Token Index: 8

Text #5

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4758. Min Act: -0.1700

Data Index: 1369407 (Python Code)

Max Activating Token Index: 5

Text #6

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4445. Min Act: -0.1700

Data Index: 1406507 (Python Code)

Max Activating Token Index: 4

Text #7

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4601. Min Act: -0.1700

Data Index: 1361666 (Python Code)

Max Activating Token Index: 5

Text #8

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4445. Min Act: -0.1700

Data Index: 1471709 (Python Code)

Max Activating Token Index: 5

Text #9

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4601. Min Act: -0.1700

Data Index: 1534829 (Python Code)

Max Activating Token Index: 4

Text #10

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4445. Min Act: -0.1700

Data Index: 1626569 (Python Code)

Max Activating Token Index: 124

Text #11

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4131. Min Act: -0.1700

Data Index: 1418388 (Python Code)

Max Activating Token Index: 387

Text #12

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4288. Min Act: -0.1700

Data Index: 1577627 (Python Code)

Max Activating Token Index: 364

Text #13

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4288. Min Act: -0.1700

Data Index: 1632559 (Python Code)

Max Activating Token Index: 4

Text #14

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4288. Min Act: -0.1700

Data Index: 1582960 (Python Code)

Max Activating Token Index: 6

Text #15

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4288. Min Act: -0.1700

Data Index: 1410347 (Python Code)

Max Activating Token Index: 142

Text #16

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4131. Min Act: -0.1700

Data Index: 1596754 (Python Code)

Max Activating Token Index: 3

Text #17

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4131. Min Act: -0.1700

Data Index: 1579213 (Python Code)

Max Activating Token Index: 3

Text #18

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.4288. Min Act: -0.1700

Data Index: 1447539 (Python Code)

Max Activating Token Index: 4

Text #19

Max Range: 3.5698. Min Range: -3.5698

Max Act: 3.3974. Min Act: -0.1700

Data Index: 1410402 (Python Code)

Max Activating Token Index: 4