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

    Neuronpedia

    INDEX
    Explanations

    overly positive

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Unique
    0.43
     Unique
    0.42
    unique
    0.40
    独特
    0.40
     unique
    0.39
     uniques
    0.39
    骓
    0.37
    獨
    0.36
     독
    0.36
     Ecc
    0.36
    POSITIVE LOGITS
     idyllic
    1.72
     cheery
    1.66
     cheerful
    1.63
     pleasant
    1.52
     blissful
    1.41
     optimistic
    1.34
     optimism
    1.30
     upbeat
    1.30
     cheerfully
    1.30
     sunny
    1.29
    Act Density 0.127%

    No Known Activations