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

    Neuronpedia

    INDEX
    Explanations

    how much or what potential

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Dou
    0.40
     Wilson
    0.40
     Kle
    0.39
     Mark
    0.39
     Ar
    0.39
     RO
    0.38
     Engagement
    0.38
     Moore
    0.38
     Landmark
    0.38
     D
    0.37
    POSITIVE LOGITS
     devient
    0.42
     ممكن
    0.42
    差异
    0.41
    possibleTypes
    0.41
    СТИ
    0.40
     deviennent
    0.40
    變得
    0.40
    linha
    0.39
    潜力
    0.39
     vraiment
    0.39
    Act Density 0.001%

    No Known Activations