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

    Neuronpedia

    INDEX
    Explanations

    mathematical transformations

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    各种
    -0.08
    -
    -0.07
     -
    -0.07
     наполн
    -0.07
    .must
    -0.07
    及
    -0.07
     linger
    -0.07
     various
    -0.07
    以及
    -0.06
     orgulho
    -0.06
    POSITIVE LOGITS
     treating
    0.13
    Treat
    0.12
     treated
    0.12
    Adjusted
    0.12
     treat
    0.12
     rewritten
    0.11
     redef
    0.11
    virtual
    0.11
     Treat
    0.11
     versione
    0.11
    Act Density 0.069%

    No Known Activations