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

    Neuronpedia

    INDEX
    Explanations

    configuration settings keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    withId
    0.46
     выбор
    0.43
     dise
    0.43
     choix
    0.41
     choice
    0.39
     aprovechar
    0.38
     zatím
    0.38
     could
    0.38
    ineux
    0.38
     thickness
    0.37
    POSITIVE LOGITS
    Disable
    0.55
    Max
    0.55
     параметр
    0.52
     параметра
    0.50
     Allow
    0.50
     directive
    0.49
    Directive
    0.49
    Maximum
    0.49
    ipv
    0.49
    Allow
    0.48
    Act Density 0.021%

    No Known Activations