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

    Neuronpedia

    INDEX
    Explanations

    holy grail

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    oscope
    0.44
    toupper
    0.43
    ườ
    0.43
    ningar
    0.43
    爷爷
    0.42
     Grandpa
    0.41
     fenómeno
    0.40
     Hoping
    0.40
    ningarna
    0.40
    ùm
    0.39
    POSITIVE LOGITS
     fault
    0.41
    с
    0.41
    //
    0.40
     purposes
    0.40
     digestible
    0.40
    ぢ
    0.40
     validity
    0.40
     date
    0.40
     dimensionless
    0.40
    بد
    0.40
    Act Density 0.001%

    No Known Activations