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

    Neuronpedia

    INDEX
    Explanations

    boolean comparisons

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric literals (integer or floating‐point constants).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     specular
    -0.07
    -du
    -0.06
    _Play
    -0.06
    ✔
    -0.06
    τικο
    -0.06
    ,↵↵
    -0.06
     vib
    -0.06
     золот
    -0.06
     Sirius
    -0.06
    आर
    -0.06
    POSITIVE LOGITS
    ree
    0.07
     Thousands
    0.07
     ==(
    0.06
     Patch
    0.06
     ла
    0.06
    .contains
    0.06
     reimburse
    0.06
    Exists
    0.06
    .compareTo
    0.06
     متن
    0.06
    Act Density 0.017%

    No Known Activations