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

    Neuronpedia

    INDEX
    Explanations

    adam

    np_max-act · gemini-2.0-flash

    This neuron never activates on any token—in effect, it’s a dead neuron that doesn’t detect any pattern.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    @implementation
    -0.07
     Bout
    -0.07
    phoneNumber
    -0.06
    363
    -0.06
     Mvc
    -0.06
    bí
    -0.06
     pager
    -0.06
     महत
    -0.06
    Tele
    -0.06
    }</
    -0.06
    POSITIVE LOGITS
     svensk
    0.07
    -esque
    0.07
     vody
    0.06
    raphics
    0.06
     فيه
    0.06
     á
    0.06
     대한민국
    0.06
    _specific
    0.06
     Bosnia
    0.06
    ओ
    0.06
    Act Density 0.005%

    No Known Activations