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

    Neuronpedia

    INDEX
    Explanations

    presenting research actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    どんどん
    0.43
     magari
    0.43
    がたくさん
    0.41
     telling
    0.40
     выход
    0.40
     forcément
    0.39
     vraiment
    0.39
     Кстати
    0.39
    Basically
    0.39
     obviamente
    0.39
    POSITIVE LOGITS
     demonstrate
    0.79
     demonstrated
    0.73
     discuss
    0.71
     presented
    0.67
     demonstrates
    0.67
     Demonstrate
    0.63
    discuss
    0.61
     propose
    0.61
     investigate
    0.59
     discusses
    0.59
    Act Density 0.010%

    No Known Activations