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

    Neuronpedia

    INDEX
    Explanations

    grammatical forms related to verb usage and sentence structure

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

    starting from

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     myſelf
    -0.62
     itſelf
    -0.62
     themſelves
    -0.57
     étoit
    -0.56
     ſche
    -0.52
     pleaſure
    -0.52
     pouvoit
    -0.50
     plufieurs
    -0.49
     rodríguez
    -0.49
     feroit
    -0.49
    POSITIVE LOGITS
     partir
    2.06
     partire
    1.41
    partir
    1.01
     starting
    0.93
     vanaf
    0.93
     from
    0.89
     Starting
    0.82
     From
    0.81
    starting
    0.80
    Starting
    0.79
    Act Density 0.001%

    No Known Activations