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

    Neuronpedia

    INDEX
    Explanations

    using for dividing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ian
    0.47
     relazioni
    0.42
     legitt
    0.42
     manifestaciones
    0.42
     italien
    0.41
     malattia
    0.40
     italiana
    0.39
     cercano
    0.39
    cienza
    0.38
     comm
    0.38
    POSITIVE LOGITS
    Surely
    0.52
     Surely
    0.50
    explo
    0.50
     sfr
    0.50
     explo
    0.45
     surely
    0.44
    Также
    0.43
    allows
    0.43
    最適
    0.42
     pozwoli
    0.42
    Act Density 0.006%

    No Known Activations