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

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

Neuron 1031 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.9000. Min Act: -0.1700

Data Index: 1575784 (Python Code)

Max Activating Token Index: 702

Text #1

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.8531. Min Act: -0.1700

Data Index: 1473851 (Python Code)

Max Activating Token Index: 128

Text #2

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.8062. Min Act: -0.1700

Data Index: 1198824 (C4 (Web Text))

Max Activating Token Index: 18

Text #3

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.7593. Min Act: -0.1700

Data Index: 1630827 (Python Code)

Max Activating Token Index: 1002

Text #4

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.7593. Min Act: -0.1700

Data Index: 1570442 (Python Code)

Max Activating Token Index: 404

Text #5

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.7436. Min Act: -0.1700

Data Index: 1086032 (C4 (Web Text))

Max Activating Token Index: 890

Text #6

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.7280. Min Act: -0.1700

Data Index: 1385540 (Python Code)

Max Activating Token Index: 671

Text #7

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.7436. Min Act: -0.1700

Data Index: 1601811 (Python Code)

Max Activating Token Index: 279

Text #8

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6810. Min Act: -0.1700

Data Index: 1433464 (Python Code)

Max Activating Token Index: 928

Text #9

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.7123. Min Act: -0.1700

Data Index: 451128 (C4 (Web Text))

Max Activating Token Index: 98

Text #10

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6967. Min Act: -0.1700

Data Index: 1598743 (Python Code)

Max Activating Token Index: 706

Text #11

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6810. Min Act: -0.1700

Data Index: 1590234 (Python Code)

Max Activating Token Index: 333

Text #12

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6497. Min Act: -0.1700

Data Index: 1517830 (Python Code)

Max Activating Token Index: 909

Text #13

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6810. Min Act: -0.1700

Data Index: 709720 (C4 (Web Text))

Max Activating Token Index: 61

Text #14

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6497. Min Act: -0.1700

Data Index: 1633259 (Python Code)

Max Activating Token Index: 4

Text #15

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6497. Min Act: -0.1700

Data Index: 998661 (C4 (Web Text))

Max Activating Token Index: 887

Text #16

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6497. Min Act: -0.1700

Data Index: 1169570 (C4 (Web Text))

Max Activating Token Index: 490

Text #17

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6810. Min Act: -0.1700

Data Index: 1596605 (Python Code)

Max Activating Token Index: 987

Text #18

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6184. Min Act: -0.1700

Data Index: 1473103 (Python Code)

Max Activating Token Index: 33

Text #19

Max Range: 3.9000. Min Range: -3.9000

Max Act: 3.6341. Min Act: -0.1700

Data Index: 1652092 (Python Code)

Max Activating Token Index: 329