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

    Neuronpedia

    INDEX
    Explanations

    names of individuals with varying ages and locations as well as associated actions and details

    oai_token-act-pair · gpt-3.5-turbo

    instances of commas used in lists or punctuation relating to numerical data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    entimes
    -0.73
    mbuds
    -0.68
    ģ«
    -0.68
    ÃįÃį
    -0.64
     Compatibility
    -0.64
    Ł
    -0.64
    Ļ½
    -0.63
    icular
    -0.61
    inar
    -0.60
    ======
    -0.60
    POSITIVE LOGITS
     who
    0.95
     pleaded
    0.87
     whose
    0.84
     grandson
    0.82
     died
    0.77
     suffers
    0.75
     aka
    0.75
     whom
    0.72
     hanged
    0.72
     granddaughter
    0.72
    Act Density 0.089%

    No Known Activations