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

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

Neuron 22 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: 5.5610. Min Range: -5.5610

Max Act: 5.5610. Min Act: -0.1700

Data Index: 283037 (C4 (Web Text))

Max Activating Token Index: 944

Text #1

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.3735. Min Act: -0.1700

Data Index: 476586 (C4 (Web Text))

Max Activating Token Index: 35

Text #2

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.3110. Min Act: -0.1700

Data Index: 91247 (C4 (Web Text))

Max Activating Token Index: 64

Text #3

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2798. Min Act: -0.1700

Data Index: 55016 (C4 (Web Text))

Max Activating Token Index: 786

Text #4

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2798. Min Act: -0.1700

Data Index: 585336 (C4 (Web Text))

Max Activating Token Index: 8

Text #5

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2485. Min Act: -0.1700

Data Index: 980793 (C4 (Web Text))

Max Activating Token Index: 855

Text #6

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2485. Min Act: -0.1700

Data Index: 123024 (C4 (Web Text))

Max Activating Token Index: 25

Text #7

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2798. Min Act: -0.1700

Data Index: 960290 (C4 (Web Text))

Max Activating Token Index: 582

Text #8

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2485. Min Act: -0.1700

Data Index: 594793 (C4 (Web Text))

Max Activating Token Index: 28

Text #9

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.2173. Min Act: -0.1700

Data Index: 457900 (C4 (Web Text))

Max Activating Token Index: 602

Text #10

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1860. Min Act: -0.1700

Data Index: 3617 (C4 (Web Text))

Max Activating Token Index: 350

Text #11

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1235. Min Act: -0.1700

Data Index: 1286678 (C4 (Web Text))

Max Activating Token Index: 533

Text #12

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.0923. Min Act: -0.1700

Data Index: 246644 (C4 (Web Text))

Max Activating Token Index: 554

Text #13

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1548. Min Act: -0.1700

Data Index: 384130 (C4 (Web Text))

Max Activating Token Index: 822

Text #14

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1860. Min Act: -0.1700

Data Index: 770790 (C4 (Web Text))

Max Activating Token Index: 19

Text #15

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1548. Min Act: -0.1700

Data Index: 700903 (C4 (Web Text))

Max Activating Token Index: 666

Text #16

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1235. Min Act: -0.1700

Data Index: 1211694 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.1235. Min Act: -0.1700

Data Index: 1018410 (C4 (Web Text))

Max Activating Token Index: 55

Text #18

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.0923. Min Act: -0.1700

Data Index: 1114431 (C4 (Web Text))

Max Activating Token Index: 16

Text #19

Max Range: 5.5610. Min Range: -5.5610

Max Act: 5.0923. Min Act: -0.1700

Data Index: 1075177 (C4 (Web Text))

Max Activating Token Index: 434