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

    Neuronpedia

    INDEX
    Explanations

    important always crucial advice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Necessity
    0.47
    Necess
    0.44
    whenever
    0.44
     necessity
    0.44
    ishing
    0.43
     Whenever
    0.43
    and
    0.43
    based
    0.42
     Necessary
    0.42
    necess
    0.42
    POSITIVE LOGITS
     baffled
    0.42
     imgs
    0.42
    焕
    0.42
    东北
    0.40
     çöze
    0.39
    ángulo
    0.38
    还没
    0.38
     setImage
    0.38
     brim
    0.38
     terminó
    0.38
    Act Density 0.010%

    No Known Activations