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

    Neuronpedia

    INDEX
    Explanations

    extracting insights and findings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Criteria
    0.42
     Um
    0.41
    criteria
    0.41
     معی
    0.41
    ssystem
    0.40
     Umgang
    0.40
    princ
    0.40
    urit
    0.39
    Criteria
    0.38
    BU
    0.38
    POSITIVE LOGITS
     insights
    0.80
     discoveries
    0.77
     conclusions
    0.72
     descobrir
    0.69
     descubrir
    0.68
     discover
    0.68
     узнать
    0.68
     investigating
    0.67
     findings
    0.66
    insights
    0.66
    Act Density 0.099%

    No Known Activations