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

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

Neuron 2046 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.5037. Min Range: -5.5037

Max Act: 5.5037. Min Act: -0.1700

Data Index: 1502577 (Python Code)

Max Activating Token Index: 839

Text #1

Max Range: 5.5037. Min Range: -5.5037

Max Act: 5.1599. Min Act: -0.1700

Data Index: 1555262 (Python Code)

Max Activating Token Index: 225

Text #2

Max Range: 5.5037. Min Range: -5.5037

Max Act: 5.1599. Min Act: -0.1700

Data Index: 1478842 (Python Code)

Max Activating Token Index: 500

Text #3

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.9724. Min Act: -0.1700

Data Index: 364066 (C4 (Web Text))

Max Activating Token Index: 1013

Text #4

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.9724. Min Act: -0.1700

Data Index: 1201231 (C4 (Web Text))

Max Activating Token Index: 144

Text #5

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.9412. Min Act: -0.1700

Data Index: 1647367 (Python Code)

Max Activating Token Index: 397

Text #6

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.9724. Min Act: -0.1700

Data Index: 1092339 (C4 (Web Text))

Max Activating Token Index: 1011

Text #7

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8474. Min Act: -0.1700

Data Index: 1117623 (C4 (Web Text))

Max Activating Token Index: 943

Text #8

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.9412. Min Act: -0.1700

Data Index: 882258 (C4 (Web Text))

Max Activating Token Index: 832

Text #9

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8787. Min Act: -0.1700

Data Index: 64590 (C4 (Web Text))

Max Activating Token Index: 859

Text #10

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8787. Min Act: -0.1700

Data Index: 1104983 (C4 (Web Text))

Max Activating Token Index: 31

Text #11

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.7537. Min Act: -0.1700

Data Index: 1142812 (C4 (Web Text))

Max Activating Token Index: 992

Text #12

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8474. Min Act: -0.1700

Data Index: 1267649 (C4 (Web Text))

Max Activating Token Index: 78

Text #13

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.9099. Min Act: -0.1700

Data Index: 167392 (C4 (Web Text))

Max Activating Token Index: 366

Text #14

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.7849. Min Act: -0.1700

Data Index: 814025 (C4 (Web Text))

Max Activating Token Index: 907

Text #15

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8162. Min Act: -0.1700

Data Index: 981767 (C4 (Web Text))

Max Activating Token Index: 768

Text #16

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.7849. Min Act: -0.1700

Data Index: 422974 (C4 (Web Text))

Max Activating Token Index: 311

Text #17

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8474. Min Act: -0.1700

Data Index: 750452 (C4 (Web Text))

Max Activating Token Index: 125

Text #18

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.8162. Min Act: -0.1700

Data Index: 483318 (C4 (Web Text))

Max Activating Token Index: 515

Text #19

Max Range: 5.5037. Min Range: -5.5037

Max Act: 4.7537. Min Act: -0.1700

Data Index: 850942 (C4 (Web Text))

Max Activating Token Index: 837