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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating the status, progress, or caution related to ongoing projects or processes

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    illet
    -0.61
    riot
    -0.56
    IVERS
    -0.56
    ukong
    -0.55
     Benefits
    -0.55
     TWO
    -0.54
    adelphia
    -0.54
     spons
    -0.53
     HELL
    -0.53
     murd
    -0.53
    POSITIVE LOGITS
     unsure
    0.94
     cautioned
    0.92
     unclear
    0.92
     uncertain
    0.89
     unlikely
    0.88
     uncertainties
    0.87
     caution
    0.86
     speculative
    0.86
     remains
    0.85
     caveats
    0.85
    Act Density 0.544%

    No Known Activations