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

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

Neuron 345 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 5.1120. Min Range: -5.1120

Max Act: 5.1120. Min Act: -0.1700

Data Index: 586219 (C4 (Web Text))

Max Activating Token Index: 607

Text #1

Max Range: 5.1120. Min Range: -5.1120

Max Act: 5.0495. Min Act: -0.1700

Data Index: 979857 (C4 (Web Text))

Max Activating Token Index: 141

Text #2

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.8933. Min Act: -0.1700

Data Index: 1556823 (Python Code)

Max Activating Token Index: 769

Text #3

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.9558. Min Act: -0.1700

Data Index: 1057508 (C4 (Web Text))

Max Activating Token Index: 333

Text #4

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.8933. Min Act: -0.1700

Data Index: 44833 (C4 (Web Text))

Max Activating Token Index: 412

Text #5

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.8620. Min Act: -0.1700

Data Index: 1567925 (Python Code)

Max Activating Token Index: 370

Text #6

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.8620. Min Act: -0.1700

Data Index: 1435774 (Python Code)

Max Activating Token Index: 582

Text #7

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7995. Min Act: -0.1700

Data Index: 951868 (C4 (Web Text))

Max Activating Token Index: 736

Text #8

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7995. Min Act: -0.1700

Data Index: 1514687 (Python Code)

Max Activating Token Index: 235

Text #9

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.8308. Min Act: -0.1700

Data Index: 1651791 (Python Code)

Max Activating Token Index: 409

Text #10

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7995. Min Act: -0.1700

Data Index: 1595895 (Python Code)

Max Activating Token Index: 353

Text #11

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7058. Min Act: -0.1700

Data Index: 1432364 (Python Code)

Max Activating Token Index: 686

Text #12

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7683. Min Act: -0.1700

Data Index: 535329 (C4 (Web Text))

Max Activating Token Index: 667

Text #13

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7683. Min Act: -0.1700

Data Index: 1127304 (C4 (Web Text))

Max Activating Token Index: 72

Text #14

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7995. Min Act: -0.1700

Data Index: 53684 (C4 (Web Text))

Max Activating Token Index: 330

Text #15

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7370. Min Act: -0.1700

Data Index: 1597955 (Python Code)

Max Activating Token Index: 478

Text #16

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7683. Min Act: -0.1700

Data Index: 1546891 (Python Code)

Max Activating Token Index: 926

Text #17

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7683. Min Act: -0.1700

Data Index: 1439857 (Python Code)

Max Activating Token Index: 250

Text #18

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7683. Min Act: -0.1700

Data Index: 1620352 (Python Code)

Max Activating Token Index: 607

Text #19

Max Range: 5.1120. Min Range: -5.1120

Max Act: 4.7370. Min Act: -0.1700

Data Index: 194120 (C4 (Web Text))

Max Activating Token Index: 330