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

    Neuronpedia

    INDEX
    Explanations

    statements expressing personal opinions about various topics

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    idium
    -0.79
    istry
    -0.73
    assisted
    -0.67
     paced
    -0.66
     concess
    -0.65
     Mandatory
    -0.62
     bailed
    -0.61
    oiler
    -0.60
    staking
    -0.60
    cius
    -0.60
    POSITIVE LOGITS
    ãĤ¤ãĥĪ
    0.86
     myself
    0.75
    IU
    0.73
    使
    0.73
     ourselves
    0.71
    omething
    0.70
    ¶æ
    0.70
     yourself
    0.70
     oneself
    0.69
    ophile
    0.69
    Act Density 8.522%

    No Known Activations