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

    Neuronpedia

    INDEX
    Explanations

    track

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric measurement tokens (e.g. decimal values), effectively detecting numeric data in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    cle
    -0.07
     Grimm
    -0.07
    ae
    -0.07
    -member
    -0.07
     Cabinets
    -0.06
     championships
    -0.06
    “
    -0.06
    _Port
    -0.06
    flu
    -0.06
    mort
    -0.06
    POSITIVE LOGITS
    ".↵↵
    0.07
     LDAP
    0.07
     %=
    0.06
    keys
    0.06
    	ArrayList
    0.06
    аются
    0.06
     appearing
    0.06
    分别
    0.06
    тра
    0.06
    чие
    0.06
    Act Density 0.000%

    No Known Activations