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

    Neuronpedia

    INDEX
    Explanations

    refugee escaping, for extended

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ano
    0.42
    atown
    0.39
    orton
    0.39
    yarn
    0.38
     схо
    0.38
    emon
    0.38
     שלנו
    0.38
     непоня
    0.37
     celestial
    0.37
    ellant
    0.37
    POSITIVE LOGITS
     Gb
    0.40
    ْل
    0.38
    晰
    0.38
     控制
    0.37
    mbH
    0.37
     ভিত
    0.36
     вну
    0.35
     শোনা
    0.35
     حمد
    0.35
    িয়াস
    0.35
    Act Density 0.000%

    No Known Activations