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

    Neuronpedia

    INDEX
    Explanations

    modal verbs indicating obligation, permission, and future actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    brids
    -0.15
    VML
    -0.14
    ilan
    -0.14
    arella
    -0.14
    ASI
    -0.13
    IGHLIGHT
    -0.13
    really
    -0.13
    ÑģÑĮ
    -0.13
    fit
    -0.13
     really
    -0.13
    POSITIVE LOGITS
     be
    0.23
     бÑĭÑĤÑĮ
    0.18
    ãģĶãģĸ
    0.18
     бÑĥÑĤи
    0.17
    º
    0.16
     být
    0.15
     not
    0.15
     nouve
    0.15
    ?↵↵↵↵↵↵
    0.15
     being
    0.15
    Act Density 0.095%

    No Known Activations