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

    Neuronpedia

    INDEX
    Explanations

    unseen data performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    錠
    0.50
     дра
    0.49
     відкри
    0.49
    𝗰
    0.48
    țit
    0.48
    ції
    0.48
    ар
    0.47
     ję
    0.47
     trabal
    0.47
     cargas
    0.46
    POSITIVE LOGITS
    le
    0.50
    il
    0.50
    el
    0.45
    ense
    0.45
    aino
    0.43
    ig
    0.42
    ibhav
    0.41
     foresight
    0.41
    المل
    0.41
    abha
    0.41
    Act Density 0.001%

    No Known Activations