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

    Neuronpedia

    INDEX
    Explanations

    verbs indicating actions such as saying, leaving, giving, taking, coming, punishing, putting, doing, taking responsibility, discovering, consolidating, reducing, making sure, calling, placing, understanding, resting, setting, and others

    oai_token-act-pair · gpt-3.5-turbo

    imperative verbs indicating actions or obligations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rawdownloadcloneembedreportprint
    -0.65
    ItemTracker
    -0.61
     Nau
    -0.59
     Duff
    -0.59
     Vermont
    -0.58
     Bruins
    -0.58
     wraps
    -0.57
    Enlarge
    -0.57
    RPG
    -0.54
     wont
    -0.54
    POSITIVE LOGITS
     oneself
    1.14
     RELEASE
    0.96
     Ratio
    0.66
    igl
    0.66
    uckland
    0.65
     yourself
    0.64
    itate
    0.63
    etheless
    0.63
     alternate
    0.63
    itime
    0.62
    Act Density 0.300%

    No Known Activations