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

    Neuronpedia

    INDEX
    Explanations

    legal licenses and agreements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     a
    0.61
    s
    0.57
     lossless
    0.55
     inhomogeneity
    0.52
    0.52
     pedir
    0.51
     havoc
    0.51
     presupuesto
    0.50
    各种
    0.48
     the
    0.48
    POSITIVE LOGITS
    ună
    0.60
     andRow
    0.60
    ™.
    0.59
    ׃
    0.57
    ەند
    0.56
    .。
    0.56
    션을
    0.56
    ™,
    0.55
    uniti
    0.55
    मूर्ति
    0.55
    Act Density 0.002%

    No Known Activations