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

    Neuronpedia

    INDEX
    Explanations

    verb conjugations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Additionally
    0.83
     Adding
    0.82
     However
    0.81
    [
    0.80
     While
    0.80
     Plant
    0.80
    Additionally
    0.78
     however
    0.77
     Specific
    0.77
     Secondly
    0.77
    POSITIVE LOGITS
     blijft
    0.86
     trebuie
    0.86
     vuelve
    0.85
     vuelva
    0.84
     évent
    0.84
     podamos
    0.84
     écoul
    0.84
     restit
    0.83
     poderá
    0.82
     estamos
    0.81
    Act Density 0.341%

    No Known Activations