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

    Neuronpedia

    INDEX
    Explanations

    needing more to achieve effect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inductance
    0.41
     dividing
    0.40
    정
    0.40
    這是
    0.40
     consolid
    0.40
     divisão
    0.39
     wherein
    0.39
    algèbre
    0.39
    pheromone
    0.39
     shrinkage
    0.38
    POSITIVE LOGITS
     account
    0.40
     workaround
    0.40
    cm
    0.39
     nowadays
    0.39
    cp
    0.39
     учетом
    0.38
     compensate
    0.38
    iliary
    0.37
    ungsver
    0.36
    flink
    0.36
    Act Density 0.003%

    No Known Activations