Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    defining base classes or models

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gods
    0.40
     elaboration
    0.38
     elaborate
    0.38
    BLY
    0.37
     nationalists
    0.37
     prelude
    0.37
     Gods
    0.37
     devils
    0.36
    eils
    0.36
     dissidents
    0.35
    POSITIVE LOGITS
     Owner
    0.56
     Founder
    0.50
     Creator
    0.49
     Perfect
    0.48
    Owner
    0.47
     UserModel
    0.46
     Model
    0.45
     Designer
    0.45
     Constructor
    0.45
    Founder
    0.45
    Act Density 0.016%

    No Known Activations