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

    Neuronpedia

    INDEX
    Explanations

    interested in or open to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     usefulness
    0.87
     useful
    0.85
     formality
    0.82
    ために
    0.77
     unusable
    0.77
     shameful
    0.75
     muffled
    0.74
     fraught
    0.74
     desirability
    0.73
     elegantly
    0.73
    POSITIVE LOGITS
     enough
    1.33
    ness
    1.19
    Enough
    1.11
    enough
    1.06
     আছেন
    1.03
    NESS
    1.03
     Enough
    1.02
    nesses
    0.94
     menerima
    0.86
     genug
    0.84
    Act Density 0.269%

    No Known Activations