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

    Neuronpedia

    INDEX
    Explanations

    core of the solution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     overall
    0.56
    Overall
    0.52
    overall
    0.52
     totale
    0.51
     sorts
    0.50
     cuidados
    0.49
    整体
    0.49
     total
    0.48
    总体
    0.46
    total
    0.44
    POSITIVE LOGITS
     matter
    0.75
    matter
    0.66
    Matter
    0.65
     Matter
    0.65
     reason
    0.61
     magic
    0.61
     이유는
    0.60
     MATTER
    0.54
    理由は
    0.52
    reason
    0.50
    Act Density 0.014%

    No Known Activations