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

    Neuronpedia

    INDEX
    Explanations

    specific measures and criteria related to evaluation

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

    Quantifiers like "only," "least," "one," "two," "many."

    np_acts-logits-general · gemini-2.0-flash

    only one or least amount

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     متعلقه
    -0.70
    참고
    -0.69
     Tripura
    -0.61
    StructEnd
    -0.60
     saites
    -0.58
    AddField
    -0.58
    HasAnnotation
    -0.57
    latego
    -0.55
     onCancelled
    -0.55
    Искәрмәләр
    -0.55
    POSITIVE LOGITS
     many
    0.66
     bezeichneter
    0.66
     another
    0.61
     some
    0.59
     one
    0.57
     fewer
    0.55
    Enllaces
    0.54
     two
    0.54
     เหล
    0.54
     several
    0.53
    Act Density 0.378%

    No Known Activations