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

    Neuronpedia

    INDEX
    Explanations

    "I" after punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Additionally
    0.49
     آنچه
    0.47
     Additionally
    0.46
    加えて
    0.44
    Certainly
    0.44
    Conversely
    0.44
     obstante
    0.42
     ceea
    0.41
     Certainly
    0.39
     ciò
    0.39
    POSITIVE LOGITS
     they
    0.56
     haven
    0.56
     everytime
    0.55
     someone
    0.53
     hadn
    0.53
     somebody
    0.51
     THEY
    0.51
     couldn
    0.49
    someone
    0.49
     wouldn
    0.49
    Act Density 0.010%

    No Known Activations