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

    Neuronpedia

    INDEX
    Explanations

    starts or ends turns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    чем
    0.75
     if
    0.73
     että
    0.73
    </blockquote>
    0.71
     it
    0.70
     том
    0.70
     If
    0.68
     jonka
    0.68
    確か
    0.66
     которое
    0.64
    POSITIVE LOGITS
     polymorphic
    0.81
    Based
    0.77
    Lens
    0.76
     jeweiligen
    0.74
     গ্রেপ্তার
    0.73
     분석
    0.73
    pdbonly
    0.72
     reacts
    0.71
    傩
    0.71
     individuell
    0.71
    Act Density 0.229%

    No Known Activations