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

    Neuronpedia

    INDEX
    Explanations

    kind of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dode
    0.40
     flakes
    0.39
    اء
    0.39
     male
    0.38
     uncompromising
    0.38
     SAW
    0.38
     largest
    0.37
     FLE
    0.37
     clones
    0.37
     modern
    0.37
    POSITIVE LOGITS
     таком
    0.44
     যখন
    0.41
    hearted
    0.41
     которой
    0.41
     котором
    0.41
    +#+#+
    0.40
    ότη
    0.40
     जेव्हा
    0.40
     जब
    0.39
    此时
    0.39
    Act Density 0.002%

    No Known Activations