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

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

Neuron 1771 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: 3.9763. Min Range: -3.9763

Max Act: 3.9763. Min Act: -0.1700

Data Index: 153640 (C4 (Web Text))

Max Activating Token Index: 37

Text #1

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.9137. Min Act: -0.1700

Data Index: 1187964 (C4 (Web Text))

Max Activating Token Index: 759

Text #2

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.9450. Min Act: -0.1700

Data Index: 511953 (C4 (Web Text))

Max Activating Token Index: 893

Text #3

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.8825. Min Act: -0.1700

Data Index: 632705 (C4 (Web Text))

Max Activating Token Index: 836

Text #4

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.9137. Min Act: -0.1700

Data Index: 1251324 (C4 (Web Text))

Max Activating Token Index: 590

Text #5

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.8825. Min Act: -0.1700

Data Index: 572057 (C4 (Web Text))

Max Activating Token Index: 789

Text #6

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.8512. Min Act: -0.1700

Data Index: 1529973 (Python Code)

Max Activating Token Index: 268

Text #7

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7886. Min Act: -0.1700

Data Index: 1153189 (C4 (Web Text))

Max Activating Token Index: 544

Text #8

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7886. Min Act: -0.1700

Data Index: 685136 (C4 (Web Text))

Max Activating Token Index: 995

Text #9

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7886. Min Act: -0.1700

Data Index: 1421547 (Python Code)

Max Activating Token Index: 710

Text #10

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7886. Min Act: -0.1700

Data Index: 1491428 (Python Code)

Max Activating Token Index: 637

Text #11

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.8199. Min Act: -0.1700

Data Index: 1638207 (Python Code)

Max Activating Token Index: 662

Text #12

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7886. Min Act: -0.1700

Data Index: 655329 (C4 (Web Text))

Max Activating Token Index: 852

Text #13

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7573. Min Act: -0.1700

Data Index: 1557080 (Python Code)

Max Activating Token Index: 260

Text #14

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7260. Min Act: -0.1700

Data Index: 1287547 (C4 (Web Text))

Max Activating Token Index: 756

Text #15

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7260. Min Act: -0.1700

Data Index: 825368 (C4 (Web Text))

Max Activating Token Index: 622

Text #16

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.7573. Min Act: -0.1700

Data Index: 336414 (C4 (Web Text))

Max Activating Token Index: 408

Text #17

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.6635. Min Act: -0.1700

Data Index: 1154552 (C4 (Web Text))

Max Activating Token Index: 268

Text #18

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.6948. Min Act: -0.1700

Data Index: 1364401 (Python Code)

Max Activating Token Index: 997

Text #19

Max Range: 3.9763. Min Range: -3.9763

Max Act: 3.6948. Min Act: -0.1700

Data Index: 1512893 (Python Code)

Max Activating Token Index: 572