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

    Neuronpedia

    INDEX
    Explanations

    names of individuals

    oai_token-act-pair · gpt-3.5-turbo

    commas and their related context in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    interstitial
    -0.79
    ancial
    -0.77
    attribute
    -0.74
    units
    -0.71
    olves
    -0.69
    ror
    -0.69
    itational
    -0.69
    icultural
    -0.67
    overs
    -0.67
    atching
    -0.66
    POSITIVE LOGITS
     meanwhile
    0.97
     Herrera
    0.95
     who
    0.93
     Duchess
    0.88
     Colo
    0.87
     whom
    0.85
     Mesh
    0.84
    nee
    0.84
    who
    0.83
     Wash
    0.82
    Act Density 0.115%

    No Known Activations