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

    Neuronpedia

    INDEX
    Explanations

    specific and crucial

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     적극
    0.75
     harsher
    0.73
    完美的
    0.72
    更高的
    0.70
     bolder
    0.69
    最適な
    0.69
     weakest
    0.69
     inappropri
    0.68
    合理的
    0.68
    最重要的
    0.68
    POSITIVE LOGITS
     ​​
    0.79
    <0xC2>
    0.77
     ­
    0.71
     ​
    0.58
     However
    0.56
    ️
    0.55
     Ð
    0.54
    ै
    0.53
     देशा
    0.53
      
    0.53
    Act Density 1.227%

    No Known Activations