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

    Neuronpedia

    INDEX
    Explanations

    phrases that signal a comparison or similarity between two different subjects

    oai_token-act-pair · gpt-3.5-turbo

    phrases that introduce comparisons or similarities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     squash
    -0.71
     Ventura
    -0.64
     bang
    -0.60
     Bild
    -0.60
     Derby
    -0.59
     Dota
    -0.58
     Stockholm
    -0.58
    $.
    -0.58
     RIS
    -0.58
     Rumble
    -0.57
    POSITIVE LOGITS
    quartered
    0.89
    wise
    0.80
    æ©Ł
    0.80
    hester
    0.77
    ctr
    0.73
    chart
    0.73
     minded
    0.73
    forward
    0.72
    wcsstore
    0.70
    etheless
    0.69
    Act Density 0.018%

    No Known Activations