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

    Neuronpedia

    INDEX
    Explanations

    code and numerical validation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Prometheus
    -0.08
     Lat
    -0.06
     lidé
    -0.06
    精
    -0.06
     disparate
    -0.06
    おり
    -0.06
     Sob
    -0.06
     경험
    -0.06
     manageable
    -0.06
    らの
    -0.06
    POSITIVE LOGITS
    umerator
    0.07
    UEST
    0.07
    autoload
    0.07
    	pid
    0.07
    신청
    0.06
    _pal
    0.06
    ↵				↵
    0.06
    created
    0.06
    	board
    0.06
    ไซ
    0.06
    Act Density 0.025%

    No Known Activations