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

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

Neuron 1958 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('gelu-3l')

Text #0

Max Range: 4.0999. Min Range: -4.0999

Max Act: 4.0999. Min Act: -0.1700

Data Index: 594398 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 4.0999. Min Range: -4.0999

Max Act: 4.0374. Min Act: -0.1700

Data Index: 192784 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 4.0999. Min Range: -4.0999

Max Act: 4.0374. Min Act: -0.1700

Data Index: 274280 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 4.0999. Min Range: -4.0999

Max Act: 4.0061. Min Act: -0.1700

Data Index: 58354 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.9749. Min Act: -0.1700

Data Index: 1161624 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8811. Min Act: -0.1700

Data Index: 1182081 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8811. Min Act: -0.1700

Data Index: 43630 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8498. Min Act: -0.1700

Data Index: 224063 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8185. Min Act: -0.1700

Data Index: 1087422 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8498. Min Act: -0.1700

Data Index: 747673 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8185. Min Act: -0.1700

Data Index: 130572 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7872. Min Act: -0.1700

Data Index: 40033 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8185. Min Act: -0.1700

Data Index: 1252659 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.8185. Min Act: -0.1700

Data Index: 707178 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7872. Min Act: -0.1700

Data Index: 367488 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7716. Min Act: -0.1700

Data Index: 816548 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7403. Min Act: -0.1700

Data Index: 1130914 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7559. Min Act: -0.1700

Data Index: 905105 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7403. Min Act: -0.1700

Data Index: 1197089 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 4.0999. Min Range: -4.0999

Max Act: 3.7559. Min Act: -0.1700

Data Index: 398616 (C4 (Web Text))

Max Activating Token Index: 1023