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

    Neuronpedia

    INDEX
    Explanations

    packages, plans, abilities, solutions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ovviamente
    0.42
     obviously
    0.41
     conseguir
    0.41
    ਤਾ
    0.38
     assolutamente
    0.37
    આત
    0.37
     integrante
    0.37
     conceding
    0.37
    ulously
    0.37
     atrocities
    0.37
    POSITIVE LOGITS
    Explained
    0.37
    误差
    0.36
    的结果
    0.35
    Importance
    0.33
     Explained
    0.33
     szület
    0.33
    不够
    0.33
     Revisited
    0.32
     Importance
    0.32
     মান
    0.32
    Act Density 0.006%

    No Known Activations