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

    Neuronpedia

    INDEX
    Explanations

    following introductory phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ถ้า
    0.89
     lots
    0.77
    ถ้า
    0.77
     usually
    0.73
     خیلی
    0.73
     trying
    0.73
     sometimes
    0.72
     אבל
    0.71
     kalau
    0.71
     things
    0.70
    POSITIVE LOGITS
     Through
    0.96
    Through
    0.88
     Specifically
    0.88
     Utilizing
    0.86
    Specifically
    0.84
     Currently
    0.82
     Notably
    0.80
     Following
    0.79
     Leveraging
    0.76
     Furthermore
    0.76
    Act Density 0.064%

    No Known Activations