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

    Neuronpedia

    INDEX
    Explanations

    dialog initiators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    FromBody
    0.69
    یدر
    0.67
     endem
    0.67
    aneously
    0.66
     bialgebras
    0.64
    яз
    0.64
    atly
    0.63
    দেশীয়
    0.63
     форму
    0.62
     appendages
    0.60
    POSITIVE LOGITS
     Hello
    1.54
     Hey
    1.50
     Thank
    1.43
    Hello
    1.42
     Sorry
    1.38
     thank
    1.37
     Yeah
    1.34
    Thank
    1.33
    Hey
    1.33
     hello
    1.31
    Act Density 0.187%

    No Known Activations