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

    Neuronpedia

    INDEX
    Explanations

    reduces, redirection, redefinition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Gruß
    0.50
     partly
    0.49
     almost
    0.48
    моль
    0.48
     Almost
    0.47
     editor
    0.46
    ankan
    0.44
     grounds
    0.44
     Fetch
    0.44
     insufficiently
    0.44
    POSITIVE LOGITS
    undant
    0.87
    uct
    0.87
    uces
    0.87
    ucing
    0.85
    uctive
    0.81
    irection
    0.81
    ressing
    0.80
    uc
    0.80
    eterm
    0.79
    efined
    0.78
    Act Density 0.094%

    No Known Activations