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

    Neuronpedia

    INDEX
    Explanations

    first half

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ‌്
    0.50
    analy
    0.47
    ಳ್
    0.46
     deste
    0.46
    γέν
    0.45
    yses
    0.45
    r
    0.45
     将
    0.45
    avacanam
    0.45
    asets
    0.44
    POSITIVE LOGITS
     unrealistic
    0.43
    答え
    0.43
    పం
    0.43
    3
    0.42
    Ư
    0.42
    Few
    0.41
     ans
    0.41
     hopeless
    0.41
     unthinkable
    0.41
    пления
    0.40
    Act Density 0.004%

    No Known Activations