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

    Neuronpedia

    INDEX
    Explanations

    model training with validation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     gifted
    -0.07
    🛐
    -0.07
     spotify
    -0.07
     опас
    -0.07
    钢琴
    -0.07
    打好
    -0.07
     formerly
    -0.07
    _mux
    -0.07
    齑
    -0.07
     Weiter
    -0.07
    POSITIVE LOGITS
    =\"$
    0.09
    र
    0.08
    utions
    0.08
    �
    0.07
     acknowledgement
    0.07
    .Image
    0.07
    北京市
    0.07
    itics
    0.07
    �
    0.07
    قول
    0.07
    Act Density 0.001%

    No Known Activations