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

    Neuronpedia

    INDEX
    Explanations

    suggestions for improvement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    partitions
    0.69
    partition
    0.67
     distributing
    0.67
    <0xAD>
    0.66
     distribuzione
    0.65
     répart
    0.64
    暌
    0.63
    ற்போது
    0.63
    arlock
    0.61
     partition
    0.61
    POSITIVE LOGITS
     improvement
    2.38
     Improvement
    2.14
    Improve
    2.12
    Improving
    2.10
     Verbesserung
    2.07
     improve
    2.07
     Improve
    2.05
     improvements
    2.03
     improving
    2.03
     Improving
    2.00
    Act Density 0.842%

    No Known Activations