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

    Neuronpedia

    INDEX
    Explanations

    specific entities or topics that are targeted or focused on

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "targeted" and its various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fo
    -0.82
    iHUD
    -0.82
    rar
    -0.80
    lycer
    -0.76
    loo
    -0.75
    batch
    -0.73
    ston
    -0.71
    SourceFile
    -0.70
    Planet
    -0.70
    heit
    -0.70
    POSITIVE LOGITS
    ted
    0.88
     squarely
    0.84
     specificity
    0.80
     targets
    0.78
     targeted
    0.74
     targeting
    0.73
     ads
    0.73
    ãĤ°
    0.70
     demographics
    0.68
     killings
    0.67
    Act Density 0.030%

    No Known Activations