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

    Neuronpedia

    INDEX
    Explanations

    the infinitive form of verbs, especially the word "to"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    continues to verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :✨
    -0.59
     confé
    -0.55
    Билгалдахарш
    -0.52
    ArgsConstructor
    -0.50
    WebControls
    -0.50
    Milwaukee
    -0.49
    AutoField
    -0.49
    urlencoded
    -0.48
     européennes
    -0.47
     مرئيه
    -0.46
    POSITIVE LOGITS
     continually
    0.69
    жнему
    0.68
    不断
    0.66
     continues
    0.66
     Contin
    0.65
    不斷
    0.65
     continuing
    0.63
    Contin
    0.63
     continual
    0.63
    不断的
    0.61
    Act Density 0.007%

    No Known Activations