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

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

Neuron 24 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: 3.7704. Min Range: -3.7704

Max Act: 3.7548. Min Act: -0.1700

Data Index: 477453 (C4 (Web Text))

Max Activating Token Index: 776

Text #1

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.7704. Min Act: -0.1700

Data Index: 322716 (C4 (Web Text))

Max Activating Token Index: 653

Text #2

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.7548. Min Act: -0.1700

Data Index: 415883 (C4 (Web Text))

Max Activating Token Index: 784

Text #3

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.7235. Min Act: -0.1700

Data Index: 1125855 (C4 (Web Text))

Max Activating Token Index: 1000

Text #4

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.7078. Min Act: -0.1700

Data Index: 1621047 (Python Code)

Max Activating Token Index: 1006

Text #5

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.7078. Min Act: -0.1700

Data Index: 429268 (C4 (Web Text))

Max Activating Token Index: 570

Text #6

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6922. Min Act: -0.1700

Data Index: 418292 (C4 (Web Text))

Max Activating Token Index: 921

Text #7

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6922. Min Act: -0.1700

Data Index: 742016 (C4 (Web Text))

Max Activating Token Index: 71

Text #8

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6922. Min Act: -0.1700

Data Index: 1357945 (C4 (Web Text))

Max Activating Token Index: 554

Text #9

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6452. Min Act: -0.1700

Data Index: 34236 (C4 (Web Text))

Max Activating Token Index: 365

Text #10

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6609. Min Act: -0.1700

Data Index: 1648181 (Python Code)

Max Activating Token Index: 807

Text #11

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6139. Min Act: -0.1700

Data Index: 47709 (C4 (Web Text))

Max Activating Token Index: 236

Text #12

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6452. Min Act: -0.1700

Data Index: 240763 (C4 (Web Text))

Max Activating Token Index: 843

Text #13

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6139. Min Act: -0.1700

Data Index: 639952 (C4 (Web Text))

Max Activating Token Index: 999

Text #14

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.5983. Min Act: -0.1700

Data Index: 1363333 (Python Code)

Max Activating Token Index: 947

Text #15

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.6139. Min Act: -0.1700

Data Index: 1404280 (Python Code)

Max Activating Token Index: 946

Text #16

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.5826. Min Act: -0.1700

Data Index: 373098 (C4 (Web Text))

Max Activating Token Index: 947

Text #17

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.5826. Min Act: -0.1700

Data Index: 334228 (C4 (Web Text))

Max Activating Token Index: 93

Text #18

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.5983. Min Act: -0.1700

Data Index: 1431691 (Python Code)

Max Activating Token Index: 857

Text #19

Max Range: 3.7704. Min Range: -3.7704

Max Act: 3.5513. Min Act: -0.1700

Data Index: 1493184 (Python Code)

Max Activating Token Index: 576