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

    Neuronpedia

    INDEX
    Explanations

    function arguments and operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <h3>
    1.64
    <eos>
    1.60
    </td>
    1.54
    <h2>
    1.41
     阅读全文
    1.40
    <start_of_image>
    1.36
    ↵↵↵↵↵↵↵↵↵↵↵
    1.34
    !<
    1.34
    ↵↵↵↵↵↵↵↵↵
    1.33
    !</
    1.31
    POSITIVE LOGITS
       
    1.11
    ;.
    1.07
    :.
    0.97
    ,.
    0.90
    -.
    0.88
    â
    0.80
    )\,
    0.78
         
    0.75
    ôs
    0.72
    ™.
    0.68
    Act Density 0.010%

    No Known Activations