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

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

Neuron 1468 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: 3.5089. Min Range: -3.5089

Max Act: 3.5089. Min Act: -0.1700

Data Index: 1155545 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.3521. Min Act: -0.1700

Data Index: 587850 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.3521. Min Act: -0.1700

Data Index: 930539 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.2578. Min Act: -0.1700

Data Index: 502838 (C4 (Web Text))

Max Activating Token Index: 157

Text #4

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.2264. Min Act: -0.1700

Data Index: 476092 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.2264. Min Act: -0.1700

Data Index: 1215711 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1635. Min Act: -0.1700

Data Index: 556689 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.3207. Min Act: -0.1700

Data Index: 1045461 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1949. Min Act: -0.1700

Data Index: 1440303 (Python Code)

Max Activating Token Index: 901

Text #9

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1949. Min Act: -0.1700

Data Index: 1599733 (Python Code)

Max Activating Token Index: 624

Text #10

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.2264. Min Act: -0.1700

Data Index: 709720 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1635. Min Act: -0.1700

Data Index: 161737 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1320. Min Act: -0.1700

Data Index: 863196 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1949. Min Act: -0.1700

Data Index: 1547931 (Python Code)

Max Activating Token Index: 559

Text #14

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1949. Min Act: -0.1700

Data Index: 528152 (C4 (Web Text))

Max Activating Token Index: 828

Text #15

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1004. Min Act: -0.1700

Data Index: 929371 (C4 (Web Text))

Max Activating Token Index: 628

Text #16

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.0689. Min Act: -0.1700

Data Index: 284437 (C4 (Web Text))

Max Activating Token Index: 649

Text #17

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1320. Min Act: -0.1700

Data Index: 1001826 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.0689. Min Act: -0.1700

Data Index: 1133396 (C4 (Web Text))

Max Activating Token Index: 758

Text #19

Max Range: 3.5089. Min Range: -3.5089

Max Act: 3.1004. Min Act: -0.1700

Data Index: 644458 (C4 (Web Text))

Max Activating Token Index: 259