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

    Neuronpedia

    INDEX
    Explanations

    closing code blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     himself
    0.69
     PERFECT
    0.65
    Break
    0.62
     myself
    0.61
    Perfect
    0.61
     ourselves
    0.58
     نفسه
    0.58
    precipitation
    0.57
     january
    0.57
    меся
    0.56
    POSITIVE LOGITS
    ()}
    0.52
    roft
    0.50
    রণে
    0.49
    oš
    0.47
    ဖို့
    0.47
    хі
    0.47
    च्या
    0.46
     Bache
    0.46
     रूप
    0.45
    ovací
    0.45
    Act Density 0.026%

    No Known Activations