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

    Neuronpedia

    INDEX
    Explanations

    refusing help while offering assistance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    €.
    0.66
     flexibility
    0.63
     rigid
    0.63
     tellement
    0.62
     individual
    0.61
     guy
    0.61
    чат
    0.59
    opian
    0.59
     शा
    0.58
     flexible
    0.57
    POSITIVE LOGITS
     pets
    0.93
     Pets
    0.86
    Pets
    0.83
     bystanders
    0.82
     dependents
    0.81
     Animals
    0.80
     animals
    0.79
    animals
    0.76
     submarines
    0.75
    aktadır
    0.73
    Act Density 0.262%

    No Known Activations