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

    Neuronpedia

    INDEX
    Explanations

    negation and excess

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     konnte
    0.29
     suited
    0.27
     suitable
    0.27
     adequate
    0.27
     compatible
    0.27
     riconosci
    0.27
     available
    0.26
     adequ
    0.26
    trust
    0.25
     pudo
    0.25
    POSITIVE LOGITS
     overly
    0.39
     excessively
    0.38
     unduly
    0.37
     necessarily
    0.36
     needlessly
    0.35
     overwhelm
    0.33
     overpower
    0.33
     unnecessarily
    0.33
     слишком
    0.32
     bogged
    0.32
    Act Density 0.101%

    No Known Activations