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

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

Neuron 748 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: 3.4077. Min Range: -3.4077

Max Act: 3.4077. Min Act: -0.1700

Data Index: 1439671 (Python Code)

Max Activating Token Index: 1023

Text #1

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.3763. Min Act: -0.1700

Data Index: 1575258 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.2507. Min Act: -0.1700

Data Index: 1517809 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0775. Min Act: -0.1700

Data Index: 98616 (C4 (Web Text))

Max Activating Token Index: 235

Text #4

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0460. Min Act: -0.1700

Data Index: 888259 (C4 (Web Text))

Max Activating Token Index: 1021

Text #5

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0617. Min Act: -0.1700

Data Index: 1567013 (Python Code)

Max Activating Token Index: 46

Text #6

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0775. Min Act: -0.1700

Data Index: 1441214 (Python Code)

Max Activating Token Index: 858

Text #7

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0460. Min Act: -0.1700

Data Index: 1596159 (Python Code)

Max Activating Token Index: 1023

Text #8

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9986. Min Act: -0.1700

Data Index: 1470286 (Python Code)

Max Activating Token Index: 313

Text #9

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0144. Min Act: -0.1700

Data Index: 1420975 (Python Code)

Max Activating Token Index: 1023

Text #10

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0144. Min Act: -0.1700

Data Index: 1484959 (Python Code)

Max Activating Token Index: 973

Text #11

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9986. Min Act: -0.1700

Data Index: 1630401 (Python Code)

Max Activating Token Index: 670

Text #12

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9986. Min Act: -0.1700

Data Index: 1559409 (Python Code)

Max Activating Token Index: 399

Text #13

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9986. Min Act: -0.1700

Data Index: 1439452 (Python Code)

Max Activating Token Index: 1023

Text #14

Max Range: 3.4077. Min Range: -3.4077

Max Act: 3.0144. Min Act: -0.1700

Data Index: 1485577 (Python Code)

Max Activating Token Index: 1023

Text #15

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9827. Min Act: -0.1700

Data Index: 1514532 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9827. Min Act: -0.1700

Data Index: 1514032 (Python Code)

Max Activating Token Index: 205

Text #17

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9827. Min Act: -0.1700

Data Index: 1537762 (Python Code)

Max Activating Token Index: 587

Text #18

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9669. Min Act: -0.1700

Data Index: 1529625 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 3.4077. Min Range: -3.4077

Max Act: 2.9827. Min Act: -0.1700

Data Index: 1449023 (Python Code)

Max Activating Token Index: 1023