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

    Neuronpedia

    INDEX
    Explanations

    high value compounds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fields
    0.72
     مشا
    0.66
    fields
    0.66
    full
    0.65
     campos
    0.63
     hints
    0.63
     full
    0.62
     zien
    0.61
     try
    0.61
     FIELDS
    0.61
    POSITIVE LOGITS
     parum
    0.90
     Compound
    0.81
     compound
    0.78
    昍
    0.78
     Protok
    0.76
    Compound
    0.75
     Việc
    0.74
    carrito
    0.74
    াতি
    0.74
     Setup
    0.74
    Act Density 0.000%

    No Known Activations