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

    Neuronpedia

    INDEX
    Explanations

    free, confidential

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    可以让
    0.43
    uestas
    0.41
     অফিসার
    0.41
     “.
    0.39
    絕對
    0.39
     "".
    0.38
     Leibn
    0.38
    ສາມາດ
    0.38
    会让
    0.38
    ا۔
    0.38
    POSITIVE LOGITS
     cris
    0.39
     confidential
    0.38
     அச
    0.38
     GOT
    0.37
     dial
    0.36
    type
    0.36
     surv
    0.36
     single
    0.35
    emic
    0.35
     armor
    0.35
    Act Density 0.004%

    No Known Activations