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

    Neuronpedia

    INDEX
    Explanations

    the word "put" along with the action that follows it

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "put" followed by contextually relevant actions or responsibilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    externalActionCode
    -0.96
    riott
    -0.66
     Austral
    -0.63
    DAQ
    -0.63
    ording
    -0.62
     Examiner
    -0.62
     Hear
    -0.61
    reader
    -0.60
    uary
    -0.60
     Magicka
    -0.60
    POSITIVE LOGITS
     together
    0.88
    ongh
    0.85
     forth
    0.85
    tin
    0.85
    rid
    0.82
    tering
    0.81
    ronics
    0.76
    tered
    0.76
    otom
    0.76
     aside
    0.75
    Act Density 0.045%

    No Known Activations