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

    Neuronpedia

    INDEX
    Explanations

    LSTM encoders and decoders

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .Health
    -0.08
     perubahan
    -0.08
     تغییر
    -0.08
     alteration
    -0.07
    zen
    -0.07
     nette
    -0.07
    ثمار
    -0.07
     zen
    -0.07
    ="#">↵
    -0.07
     unavoidable
    -0.07
    POSITIVE LOGITS
     decoder
    0.08
    decoder
    0.08
    iterations
    0.08
    itera
    0.08
     reconstruct
    0.08
    ooka
    0.07
     Oo
    0.07
     reunited
    0.07
    Decoder
    0.07
    .decoder
    0.07
    Act Density 0.001%

    No Known Activations