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

    Neuronpedia

    INDEX
    Explanations

    says

    np_max-act · gemini-2.0-flash

    The neuron specializes in spotting reporting or attribution verbs (e.g. “says,” “states,” “reads”) that introduce quoted or reported content.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    环境
    -0.07
     palate
    -0.07
    |null
    -0.06
     většinou
    -0.06
     NX
    -0.06
     제출
    -0.06
     spur
    -0.06
    吐
    -0.06
    236
    -0.06
     duct
    -0.06
    POSITIVE LOGITS
     detailing
    0.07
     reads
    0.07
    ston
    0.06
     propTypes
    0.06
    dst
    0.06
    裏
    0.06
    .MenuStrip
    0.06
    روج
    0.06
    _machine
    0.06
     sufferers
    0.06
    Act Density 0.024%

    No Known Activations