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

    Neuronpedia

    INDEX
    Explanations

    Overused or excessive terms and phrases

    oai_token-act-pair · gpt-3.5-turbo

    instances of the prefix "over" or related forms indicating excessive actions or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    senal
    -0.68
    minster
    -0.68
    ulhu
    -0.66
    sama
    -0.61
    silver
    -0.60
    Lago
    -0.59
    PLA
    -0.59
     Solitaire
    -0.58
    MpServer
    -0.58
     NCT
    -0.57
    POSITIVE LOGITS
    atch
    0.89
    agog
    0.84
    uling
    0.81
    urring
    0.81
    uled
    0.80
    adows
    0.80
    ider
    0.79
    gment
    0.79
    ases
    0.79
    uner
    0.79
    Act Density 0.074%

    No Known Activations