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

    Neuronpedia

    INDEX
    Explanations

    principles and guidelines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     allocations
    0.43
     racks
    0.42
     maple
    0.42
     float
    0.42
     allocated
    0.40
     citrus
    0.39
     （
    0.39
     sightings
    0.38
     combines
    0.38
     selectable
    0.38
    POSITIVE LOGITS
     моя
    0.56
    我不
    0.49
    Neces
    0.48
    私は
    0.47
     журнали
    0.46
     abhor
    0.45
    笔者
    0.45
    Мне
    0.44
    我也
    0.44
    我认为
    0.44
    Act Density 1.043%

    No Known Activations