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

    Neuronpedia

    INDEX
    Explanations

    hold or holding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    쑤
    -0.84
     podem
    -0.77
     freuen
    -0.76
     déplacer
    -0.73
    lagten
    -0.73
    částí
    -0.70
     ór
    -0.70
     include
    -0.69
    怅
    -0.69
    twimg
    -0.69
    POSITIVE LOGITS
     hold
    3.86
     holds
    3.84
     Holds
    3.14
    holds
    2.84
     Hold
    2.84
    hold
    2.78
    Hold
    2.78
     HOLD
    2.44
    HOLD
    2.33
     holding
    2.31
    Act Density 0.057%

    No Known Activations