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

    Neuronpedia

    INDEX
    Explanations

    circles

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ున
    -0.08
    েতে
    -0.08
    ాలో
    -0.08
     اکثر
    -0.07
     robust
    -0.07
    ురు
    -0.07
     تعامل
    -0.07
     ذ
    -0.07
     attend
    -0.07
    ిన
    -0.07
    POSITIVE LOGITS
     tiers
    0.12
    Layers
    0.11
    _layers
    0.10
     Layers
    0.10
     progressively
    0.10
     layers
    0.10
     níveis
    0.10
    gradu
    0.10
     graduating
    0.10
    tiers
    0.09
    Act Density 0.012%

    No Known Activations