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

    Neuronpedia

    INDEX
    Explanations

    moderate qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     magnificent
    0.73
     великолеп
    0.73
     brilliantly
    0.73
     brilliant
    0.72
    极致
    0.72
    极其
    0.69
     increí
    0.69
     incroyable
    0.68
     enormes
    0.68
     exquisite
    0.68
    POSITIVE LOGITS
     decent
    2.17
     reasonably
    2.17
     decently
    2.11
     reasonable
    1.88
     moderately
    1.82
     fairly
    1.75
    reasonably
    1.71
     relatively
    1.55
    reasonable
    1.54
     Reasonable
    1.54
    Act Density 0.497%

    No Known Activations