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

    Neuronpedia

    INDEX
    Explanations

    specific closing words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Act
    0.90
    Methods
    0.90
    Path
    0.83
    age
    0.82
    methods
    0.82
    І
    0.81
    only
    0.81
    As
    0.80
    ess
    0.80
    и
    0.79
    POSITIVE LOGITS
     LinkedList
    0.93
     audacious
    0.88
     cappuccino
    0.87
     guava
    0.84
     furry
    0.81
     cel
    0.80
     rebranding
    0.80
     hairy
    0.80
     monstrous
    0.80
     multitude
    0.79
    Act Density 0.001%

    No Known Activations