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

    Neuronpedia

    INDEX
    Explanations

    striking image; savages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Handle
    0.53
    じゃない
    0.49
     догово
    0.48
    いの
    0.48
     sio
    0.47
     ハート
    0.47
    Slice
    0.47
    ש
    0.47
     pariatur
    0.46
    りん
    0.46
    POSITIVE LOGITS
    </u>
    0.54
    </span>
    0.50
    ādi
    0.46
     Egypt
    0.46
    amak
    0.46
    āj
    0.46
     DF
    0.45
    āta
    0.45
    </td>
    0.45
    vald
    0.44
    Act Density 0.002%

    No Known Activations