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

    Neuronpedia

    INDEX
    Explanations

    expressions of typicality or common attributes related to various subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    the word "typical."

    np_acts-logits-general · gemini-2.0-flash

    typical descriptor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Cordero
    -0.66
    orra
    -0.63
     plais
    -0.60
    mercedes
    -0.59
     Melinda
    -0.58
    花板
    -0.58
     Injectable
    -0.57
    grunt
    -0.57
    hands
    -0.57
     hose
    -0.56
    POSITIVE LOGITS
    featureID
    0.85
     dataSnapshot
    0.80
     Merkmale
    0.80
    TagMode
    0.79
    typical
    0.78
     propOrder
    0.77
    ity
    0.77
    neros
    0.76
    windowFixed
    0.73
     Waray
    0.72
    Act Density 0.153%

    No Known Activations