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

    Neuronpedia

    INDEX
    Explanations

    Code configuration

    np_max-act · gemini-2.0-flash

    the main thing this neuron does is detect metadata field names (i.e. the keys in key–value headers) in structured files

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    /be
    -0.08
    Joy
    -0.08
     restaurant
    -0.07
     Black
    -0.07
     Bang
    -0.07
    553
    -0.07
    _on
    -0.06
    Italic
    -0.06
    _sb
    -0.06
    .album
    -0.06
    POSITIVE LOGITS
     FAA
    0.06
    ,**
    0.06
    ancellor
    0.06
    tility
    0.06
     Pompeo
    0.06
    存于
    0.06
    "]);
↵
    0.06
    اون
    0.06
     دریا
    0.06
    ()},↵
    0.06
    Act Density 0.160%

    No Known Activations