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

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

Neuron 2034 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.9414. Min Act: -0.1700

Data Index: 539298 (C4 (Web Text))

Max Activating Token Index: 715

Text #1

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.9414. Min Act: -0.1700

Data Index: 570884 (C4 (Web Text))

Max Activating Token Index: 781

Text #2

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 1088061 (C4 (Web Text))

Max Activating Token Index: 480

Text #3

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 582401 (C4 (Web Text))

Max Activating Token Index: 573

Text #4

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1535043 (Python Code)

Max Activating Token Index: 6

Text #5

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1490575 (Python Code)

Max Activating Token Index: 6

Text #6

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1583032 (Python Code)

Max Activating Token Index: 6

Text #7

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1469392 (Python Code)

Max Activating Token Index: 6

Text #8

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1584128 (Python Code)

Max Activating Token Index: 6

Text #9

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1366084 (Python Code)

Max Activating Token Index: 6

Text #10

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1599945 (Python Code)

Max Activating Token Index: 6

Text #11

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1609760 (Python Code)

Max Activating Token Index: 6

Text #12

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8163. Min Act: -0.1700

Data Index: 1449835 (Python Code)

Max Activating Token Index: 6

Text #13

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 815832 (C4 (Web Text))

Max Activating Token Index: 383

Text #14

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 1158149 (C4 (Web Text))

Max Activating Token Index: 486

Text #15

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 661548 (C4 (Web Text))

Max Activating Token Index: 279

Text #16

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 622794 (C4 (Web Text))

Max Activating Token Index: 276

Text #17

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 212218 (C4 (Web Text))

Max Activating Token Index: 561

Text #18

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 900737 (C4 (Web Text))

Max Activating Token Index: 284

Text #19

Max Range: 3.9414. Min Range: -3.9414

Max Act: 3.8475. Min Act: -0.1700

Data Index: 754952 (C4 (Web Text))

Max Activating Token Index: 630