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

    Neuronpedia

    INDEX
    Explanations

    easier, difficult, possible, important

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    dimensioni
    0.39
    нула
    0.39
    பூ
    0.38
     Preferably
    0.38
    FAQ
    0.38
     NEEDS
    0.38
    Suggestion
    0.37
    >)</
    0.37
     זאת
    0.37
     işlemler
    0.36
    POSITIVE LOGITS
     impossible
    0.70
     easier
    0.61
     imposible
    0.60
     difficult
    0.59
     unwise
    0.59
     easy
    0.58
    impossible
    0.54
     impossibile
    0.54
     wise
    0.53
     imperative
    0.53
    Act Density 0.024%

    No Known Activations