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

    Neuronpedia

    INDEX
    Explanations

    end of phrase

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.71
     Öncelikle
    0.69
     pptd
    0.69
    ]',
    0.66
     있는데요
    0.65
    LLCATS
    0.65
    二是
    0.65
    𓏧
    0.64
     얘는
    0.63
    ශ්‍ය
    0.63
    POSITIVE LOGITS
    <eos>
    2.61
    ៕
    2.14
    <start_of_image>
    1.75
    ).\\
    1.70
    </blockquote>
    1.69
    ↵↵↵↵↵
    1.67
    .").
    1.63
    ↵↵↵
    1.63
    ↵↵↵↵
    1.62
    .”
    1.59
    Act Density 1.156%

    No Known Activations