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

    Neuronpedia

    INDEX
    Explanations

    code snippet explanation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     spp
    0.42
     Siempre
    0.40
     annually
    0.39
    超過
    0.38
     always
    0.38
     время
    0.38
     time
    0.37
     Always
    0.37
    超过
    0.37
     عام
    0.37
    POSITIVE LOGITS
     indicates
    0.65
     indiquant
    0.64
     evidentemente
    0.63
     указывает
    0.62
     evidently
    0.57
     Indicates
    0.56
     indicating
    0.55
    显然
    0.55
     vermutlich
    0.54
     offenbar
    0.54
    Act Density 0.511%

    No Known Activations