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

    Neuronpedia

    INDEX
    Explanations

    right, correct, best

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Anything
    0.75
     Anything
    0.73
    ᓐ
    0.71
     provient
    0.69
    ಂದರೆ
    0.68
     utilice
    0.68
     činjen
    0.67
     вообще
    0.66
     principalmente
    0.66
    に関しては
    0.66
    POSITIVE LOGITS
     right
    3.23
     correct
    2.91
     juiste
    2.78
     Right
    2.59
    right
    2.58
    Right
    2.46
     richtige
    2.40
     RIGHT
    2.35
    correct
    2.35
    正确的
    2.31
    Act Density 0.518%

    No Known Activations