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

    Neuronpedia

    INDEX
    Explanations

    the term "proper" in various contexts

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

    correct or suitable descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    rungsseite
    -0.60
    Gweler
    -0.58
    stylers
    -0.49
     للمعارف
    -0.45
     aveug
    -0.41
     vocales
    -0.41
     azon
    -0.40
     auroit
    -0.38
     مرئيه
    -0.38
     femen
    -0.36
    POSITIVE LOGITS
     proper
    0.97
     Proper
    0.91
    Proper
    0.90
    proper
    0.90
     PROPER
    0.87
     Properly
    0.75
     properly
    0.75
     decent
    0.73
    ちゃんと
    0.72
    properly
    0.71
    Act Density 0.033%

    No Known Activations