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

    Neuronpedia

    INDEX
    Explanations

    NBA

    np_max-act · gemini-2.0-flash

    This neuron detects placeholder tokens of the form “NAME_<number>” (i.e. placeholder variable names).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    гов
    -0.06
     RCS
    -0.06
     případech
    -0.06
    -st
    -0.06
    External
    -0.06
    ٫
    -0.06
    errals
    -0.06
     Sites
    -0.06
     absorb
    -0.06
    VG
    -0.06
    POSITIVE LOGITS
     ör
    0.07
    ="#"
    0.07
     Ober
    0.06
    .loader
    0.06
     Assume
    0.06
    Sir
    0.06
    urv
    0.06
    centroid
    0.06
    PMC
    0.06
    >Date
    0.06
    Act Density 0.290%

    No Known Activations