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

    Neuronpedia

    INDEX
    Explanations

    Latin words or phrases

    oai_token-act-pair · gpt-3.5-turbo

    abstract or complex concepts and themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Luck
    -0.69
    rail
    -0.66
    eness
    -0.65
    models
    -0.65
    odder
    -0.65
    October
    -0.63
    ilateral
    -0.62
    Warren
    -0.61
    Topics
    -0.61
    icky
    -0.59
    POSITIVE LOGITS
    xit
    1.02
    llo
    0.90
    ndum
    0.88
    pta
    0.87
    lla
    0.87
    utsche
    0.84
    ller
    0.82
    lda
    0.81
    pt
    0.81
     produ
    0.78
    Act Density 0.140%

    No Known Activations