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

    Neuronpedia

    INDEX
    Explanations

    performance comparison

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    更为
    0.89
    更加
    0.86
    より
    0.83
     bardziej
    0.80
     більш
    0.80
     bolj
    0.77
    bardziej
    0.73
     보다
    0.72
    よりも
    0.70
     अपेक्षाकृत
    0.69
    POSITIVE LOGITS
     stead
    0.52
     notably
    0.44
     gent
    0.43
    marg
    0.43
     objectively
    0.41
     tast
    0.39
    stead
    0.38
     secur
    0.38
     lus
    0.38
     demonstr
    0.38
    Act Density 0.330%

    No Known Activations