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

    Neuronpedia

    INDEX
    Explanations

    starts of personal statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    uance
    0.72
     nantinya
    0.72
    aternary
    0.69
    ize
    0.68
    ematic
    0.68
    祀
    0.68
    oint
    0.67
     ask
    0.67
     shall
    0.66
     metadata
    0.66
    POSITIVE LOGITS
     unglaublich
    1.43
     excelente
    1.41
    Excelente
    1.40
     incredibly
    1.40
    Honestly
    1.40
    Bought
    1.39
     Bought
    1.37
     Excelente
    1.37
     দুর্দান্ত
    1.36
     Honestly
    1.35
    Act Density 0.098%

    No Known Activations