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

    Neuronpedia

    INDEX
    Explanations

    detections of nationalities or origins

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and phrases that link multiple items or subjects together

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ynthesis
    -0.85
     cages
    -0.72
    omnia
    -0.72
    orescence
    -0.71
    NetMessage
    -0.69
    otions
    -0.69
    uilding
    -0.68
    umbnails
    -0.68
     rigs
    -0.67
    ortex
    -0.66
    POSITIVE LOGITS
     Italian
    1.62
     German
    1.60
     Portuguese
    1.57
     Belgian
    1.56
     Vietnamese
    1.53
     Romanian
    1.53
     Hungarian
    1.53
     Taiwanese
    1.51
     Indonesian
    1.50
     Jordanian
    1.49
    Act Density 0.193%

    No Known Activations