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

    Neuronpedia

    INDEX
    Explanations

    maximize window, get url

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nights
    0.43
     guilt
    0.42
    度が
    0.40
     depression
    0.39
     driftwood
    0.39
     mant
    0.38
     mor
    0.38
     circulation
    0.37
     deterior
    0.37
     hc
    0.37
    POSITIVE LOGITS
    deleteAll
    0.68
    maximize
    0.64
    timeouts
    0.52
    implicitly
    0.52
    window
    0.51
    manage
    0.51
    Maximize
    0.51
    Deleting
    0.48
    maximizing
    0.48
     deleteAll
    0.48
    Act Density 0.000%

    No Known Activations