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

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

Neuron 1274 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: 5.0901. Min Range: -5.0901

Max Act: 5.0901. Min Act: -0.1700

Data Index: 594364 (C4 (Web Text))

Max Activating Token Index: 975

Text #1

Max Range: 5.0901. Min Range: -5.0901

Max Act: 5.0901. Min Act: -0.1700

Data Index: 1160905 (C4 (Web Text))

Max Activating Token Index: 72

Text #2

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.9651. Min Act: -0.1700

Data Index: 365657 (C4 (Web Text))

Max Activating Token Index: 733

Text #3

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.9026. Min Act: -0.1700

Data Index: 1023031 (C4 (Web Text))

Max Activating Token Index: 616

Text #4

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.8089. Min Act: -0.1700

Data Index: 1650561 (Python Code)

Max Activating Token Index: 489

Text #5

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.8089. Min Act: -0.1700

Data Index: 1217248 (C4 (Web Text))

Max Activating Token Index: 561

Text #6

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7776. Min Act: -0.1700

Data Index: 556012 (C4 (Web Text))

Max Activating Token Index: 328

Text #7

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7464. Min Act: -0.1700

Data Index: 419148 (C4 (Web Text))

Max Activating Token Index: 966

Text #8

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7464. Min Act: -0.1700

Data Index: 491501 (C4 (Web Text))

Max Activating Token Index: 589

Text #9

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.8089. Min Act: -0.1700

Data Index: 122830 (C4 (Web Text))

Max Activating Token Index: 231

Text #10

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.8089. Min Act: -0.1700

Data Index: 1170711 (C4 (Web Text))

Max Activating Token Index: 594

Text #11

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7464. Min Act: -0.1700

Data Index: 1184406 (C4 (Web Text))

Max Activating Token Index: 299

Text #12

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7776. Min Act: -0.1700

Data Index: 281195 (C4 (Web Text))

Max Activating Token Index: 789

Text #13

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7151. Min Act: -0.1700

Data Index: 247956 (C4 (Web Text))

Max Activating Token Index: 850

Text #14

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7151. Min Act: -0.1700

Data Index: 1627867 (Python Code)

Max Activating Token Index: 22

Text #15

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7151. Min Act: -0.1700

Data Index: 165197 (C4 (Web Text))

Max Activating Token Index: 226

Text #16

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7151. Min Act: -0.1700

Data Index: 143879 (C4 (Web Text))

Max Activating Token Index: 672

Text #17

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.6839. Min Act: -0.1700

Data Index: 1279252 (C4 (Web Text))

Max Activating Token Index: 518

Text #18

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.7464. Min Act: -0.1700

Data Index: 60722 (C4 (Web Text))

Max Activating Token Index: 111

Text #19

Max Range: 5.0901. Min Range: -5.0901

Max Act: 4.6839. Min Act: -0.1700

Data Index: 372942 (C4 (Web Text))

Max Activating Token Index: 698