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

    Neuronpedia

    INDEX
    Explanations

    conveying core statements or rules

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ：「
    0.46
    不想
    0.45
    はじめ
    0.44
    ，「
    0.43
    !”
    0.41
    Anything
    0.41
     afterwards
    0.41
    这也是
    0.41
     nahin
    0.41
     ఇదే
    0.41
    POSITIVE LOGITS
     essentially
    0.74
     basically
    0.71
     básicamente
    0.62
    basically
    0.62
     simply
    0.60
     basicamente
    0.58
     principally
    0.57
     Basically
    0.56
    essentially
    0.56
     simplesmente
    0.55
    Act Density 0.014%

    No Known Activations