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

    Neuronpedia

    INDEX
    Explanations

    predict

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Citizenship
    -0.07
    Tab
    -0.07
     COS
    -0.07
     zug
    -0.07
    issues
    -0.07
    [MAX
    -0.07
     giấy
    -0.07
    anlar
    -0.07
     terminology
    -0.06
     dos
    -0.06
    POSITIVE LOGITS
     predict
    0.10
     predicted
    0.10
     Predict
    0.09
     predictor
    0.08
     predicting
    0.08
     predicts
    0.08
    Predict
    0.08
     recipes
    0.08
     prediction
    0.08
    redict
    0.08
    Act Density 0.044%

    No Known Activations