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

    Neuronpedia

    INDEX
    Explanations

    td

    np_max-act · gemini-2.0-flash

    This neuron detects HTML markup tokens—i.e. the angle-bracket tag names and punctuation that mark up HTML (like `<td>`, `</tr>`, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     얘
    -0.07
    PLICATE
    -0.07
    []>↵
    -0.06
    cede
    -0.06
    	↵	↵↵
    -0.06
    ';
↵
↵
    -0.06
    seniz
    -0.06
    !*\↵
    -0.06
    READ
    -0.06
    입니다
    -0.06
    POSITIVE LOGITS
    OPS
    0.07
    grass
    0.07
     prostitute
    0.06
     democratic
    0.06
    ,set
    0.06
     Warp
    0.06
    _PROXY
    0.06
    ops
    0.06
     crawling
    0.06
     cohesion
    0.06
    Act Density 0.012%

    No Known Activations