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

    Neuronpedia

    INDEX
    Explanations

    data annotation tools

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Consistency
    0.42
    Trick
    0.41
    ಸ್ತು
    0.41
     Someone
    0.40
    刚才
    0.40
    一个人
    0.39
     trick
    0.39
     Compass
    0.38
     Trick
    0.38
    可以将
    0.38
    POSITIVE LOGITS
    massive
    0.45
     большие
    0.43
     ähn
    0.40
     apex
    0.39
    similar
    0.39
     large
    0.37
    large
    0.37
     대부분
    0.37
     behem
    0.37
     Tüm
    0.36
    Act Density 0.005%

    No Known Activations