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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing preference or suggestion

    oai_token-act-pair · gpt-3.5-turbo

    expressions of doing something effectively or satisfactorily

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hyde
    -0.82
    ategory
    -0.72
    laus
    -0.70
    heast
    -0.68
    adena
    -0.68
    EStreamFrame
    -0.64
    anos
    -0.63
     furiously
    -0.62
    mid
    -0.61
    zanne
    -0.60
    POSITIVE LOGITS
     behaved
    0.76
    Initialized
    0.71
    ector
    0.71
    NESS
    0.69
    lied
    0.68
    ogical
    0.68
    ECT
    0.68
     suited
    0.68
    iberal
    0.66
     tuned
    0.65
    Act Density 0.048%

    No Known Activations