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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron detects HTML (or XML) markup tags (e.g., elements like <div>, <pre>, <span>, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     hroz
    -0.07
    ARIANT
    -0.07
    (ps
    -0.07
    üns
    -0.07
    /signup
    -0.06
     LDS
    -0.06
    .mm
    -0.06
     Ke
    -0.06
     Об
    -0.06
    contrast
    -0.06
    POSITIVE LOGITS
    _PASSWORD
    0.06
     rasp
    0.06
    ugh
    0.06
    enance
    0.06
    omet
    0.06
    テレビ
    0.06
    دن
    0.06
    .Cmd
    0.06
    َد
    0.06
     Drinking
    0.06
    Act Density 0.014%

    No Known Activations