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

    Neuronpedia

    INDEX
    Explanations

    data types

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     наиболее
    -0.06
     siguientes
    -0.06
    above
    -0.06
    Temp
    -0.06
    حات
    -0.06
    Publish
    -0.06
     byla
    -0.06
     buddy
    -0.06
     Thoughts
    -0.06
     pronunciation
    -0.06
    POSITIVE LOGITS
    innacle
    0.07
    _npc
    0.07
     اعتر
    0.07
    .Enabled
    0.06
     Lama
    0.06
     yOffset
    0.06
    <dynamic
    0.06
     ""),↵
    0.06
    员
    0.06
     кис
    0.06
    Act Density 0.015%

    No Known Activations