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

    Neuronpedia

    INDEX
    Explanations

    offer to elaborate further

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Thanks
    0.87
     gracias
    0.83
     مشرف
    0.80
     thanks
    0.74
     zabud
    0.73
     मुहूर्त
    0.73
     grazie
    0.73
     pleine
    0.72
     ምክ
    0.72
     बेस्ट
    0.72
    POSITIVE LOGITS
     do
    0.89
    Do
    0.77
    do
    0.75
     does
    0.72
     instance
    0.72
     Do
    0.68
     reactants
    0.67
    does
    0.67
    instance
    0.64
     ===
    0.64
    Act Density 0.070%

    No Known Activations