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

    Neuronpedia

    INDEX
    Explanations

    lists and questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     except
    0.81
    াঃ
    0.80
     Except
    0.76
     Tails
    0.75
    +:
    0.74
    :\
    0.73
     exemple
    0.73
     Router
    0.71
    +,
    0.71
    যেমন
    0.70
    POSITIVE LOGITS
     mü
    0.55
    underbrace
    0.54
    ë
    0.53
    quantitative
    0.53
     aristocratic
    0.52
    ev
    0.51
    윙
    0.50
    alien
    0.50
    identity
    0.49
    administrative
    0.49
    Act Density 0.283%

    No Known Activations