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

    Neuronpedia

    INDEX
    Explanations

    obligated to follow

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    覂
    0.55
     rupp
    0.53
     appels
    0.53
     است
    0.53
    enging
    0.52
     prestación
    0.52
     années
    0.52
     ersetzt
    0.52
     festgestellt
    0.52
     جات
    0.51
    POSITIVE LOGITS
     Ratio
    0.52
    ጹ
    0.50
    좀
    0.50
     chained
    0.48
     ratio
    0.47
     commented
    0.47
    分数
    0.46
    砖
    0.45
    ר
    0.45
    稍
    0.45
    Act Density 0.000%

    No Known Activations