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

    Neuronpedia

    INDEX
    Explanations

    instruction formats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <strong>
    0.45
     saddo
    0.40
     crossroads
    0.39
     zero
    0.38
    <u>
    0.37
     et
    0.37
     [
    0.36
     zéro
    0.36
     either
    0.36
    </u>
    0.35
    POSITIVE LOGITS
     Despite
    0.56
    Despite
    0.51
     Несмотря
    0.46
    Learn
    0.45
     несмотря
    0.44
     despite
    0.44
     explores
    0.44
     узна
    0.43
    दोनों
    0.43
    介绍了
    0.43
    Act Density 0.004%

    No Known Activations