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

    Neuronpedia

    INDEX
    Explanations

    Feedback loops

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Specified
    -0.08
     beachten
    -0.08
     Unified
    -0.07
     Depending
    -0.07
     iceberg
    -0.07
    OUGH
    -0.07
    Unified
    -0.07
     preparo
    -0.07
    Cancelled
    -0.07
    ога
    -0.07
    POSITIVE LOGITS
    反馈
    0.12
     feedback
    0.12
     wiederum
    0.11
    促进
    0.11
    feedback
    0.10
    さらに
    0.10
    促
    0.10
    助
    0.10
    不断
    0.09
    Feedback
    0.09
    Act Density 0.045%

    No Known Activations