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

    Neuronpedia

    INDEX
    Explanations

    read carefully

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    への
    0.46
     usable
    0.42
    使える
    0.41
     tested
    0.40
    false
    0.39
     ไง
    0.39
     explored
    0.39
    superior
    0.38
     дальней
    0.38
    ใช้งาน
    0.38
    POSITIVE LOGITS
     attentively
    1.00
     aloud
    0.94
     carefully
    0.86
     intently
    0.84
    仔细
    0.82
     thoroughly
    0.75
     внима
    0.71
     atent
    0.68
     Carefully
    0.68
    认真
    0.64
    Act Density 0.012%

    No Known Activations