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

    Neuronpedia

    INDEX
    Explanations

    parts of documents that contain a series of asterisks '***'

    oai_token-act-pair · gpt-3.5-turbo

    symbols or special characters used for emphasis or separation in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    liness
    -0.72
    zza
    -0.67
    uces
    -0.66
    oby
    -0.65
     scattering
    -0.64
     srf
    -0.64
    kered
    -0.64
    onomy
    -0.62
     foc
    -0.62
     Rico
    -0.61
    POSITIVE LOGITS
    WARNING
    1.02
    THIS
    0.92
    UPDATE
    0.91
    edited
    0.89
    EDIT
    0.89
    NEW
    0.88
    hole
    0.86
    kw
    0.84
    NOT
    0.83
    COMPLE
    0.83
    Act Density 0.023%

    No Known Activations