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

    Neuronpedia

    INDEX
    Explanations

    comparisons or contrasts emphasizing a heightened or exceptional level of something

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing a comparison to a past state of being, usually emphasizing an increase or improvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oute
    -0.70
     Eug
    -0.65
    enko
    -0.64
     McA
    -0.64
     amy
    -0.62
    istries
    -0.62
    ouri
    -0.61
     Cumm
    -0.59
    riad
    -0.57
    LU
    -0.57
    POSITIVE LOGITS
    THING
    0.82
     dreamed
    0.76
    theless
    0.76
     imaginable
    0.74
     imagined
    0.72
     before
    0.72
     existed
    0.72
     BEFORE
    0.71
     cared
    0.71
     recorded
    0.69
    Act Density 0.019%

    No Known Activations