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

    Neuronpedia

    INDEX
    Explanations

    step-by-step explanation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    においても
    0.45
    我也
    0.44
     anyway
    0.43
     saya
    0.42
     nonetheless
    0.42
     wohl
    0.41
    我会
    0.41
    恐怕
    0.41
     nevertheless
    0.40
    我也是
    0.38
    POSITIVE LOGITS
     Importantly
    0.82
     Effectively
    0.78
     importantly
    0.71
    Essentially
    0.70
     crucially
    0.70
     Thereby
    0.70
     Essentially
    0.68
    Таким
    0.68
     Afterwards
    0.67
     এইরূপে
    0.65
    Act Density 0.032%

    No Known Activations