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

    Neuronpedia

    INDEX
    Explanations

    the conditional and subjunctive forms of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    gräns
    -0.52
     Cans
    -0.52
     assurent
    -0.47
     खरीदा
    -0.47
     offenbar
    -0.46
    alló
    -0.46
     identifica
    -0.45
     Ceramby
    -0.45
    rzej
    -0.45
     duquel
    -0.45
    POSITIVE LOGITS
     gdyby
    0.84
     незавершена
    0.83
     wouldn
    0.80
    SequentialGroup
    0.79
    would
    0.74
    wouldn
    0.72
     would
    0.71
    SpringRunner
    0.69
     Would
    0.68
    IntoConstraints
    0.68
    Act Density 0.163%

    No Known Activations