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

    Neuronpedia

    INDEX
    Explanations

    affectionate greetings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     awesome
    0.93
     espectacular
    0.86
     badass
    0.85
     assignment
    0.77
     guys
    0.76
     segmented
    0.76
     guy
    0.75
     massive
    0.74
     firepower
    0.74
     강
    0.74
    POSITIVE LOGITS
     dear
    1.19
     dearest
    0.89
     rather
    0.86
     mustn
    0.85
    Dear
    0.85
     darling
    0.84
     grie
    0.84
    oh
    0.84
    dear
    0.83
     Oh
    0.82
    Act Density 0.092%

    No Known Activations