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

    Neuronpedia

    INDEX
    Explanations

    important followed by note

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    some
    0.66
     some
    0.64
    Important
    0.64
    essence
    0.62
     Important
    0.61
     needed
    0.59
    Some
    0.59
     Some
    0.59
    part
    0.58
    innovation
    0.58
    POSITIVE LOGITS
    ientes
    1.01
     truths
    1.00
    fares
    0.99
     dosages
    0.97
     melodies
    0.95
    하거나
    0.94
    izações
    0.93
     treatises
    0.93
    ámicas
    0.93
     protrusions
    0.93
    Act Density 0.120%

    No Known Activations