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

    Neuronpedia

    INDEX
    Explanations

    continue. immediately following certain directives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Well
    0.44
    Fu
    0.41
    🥬
    0.41
    ऱ्यांना
    0.40
    But
    0.39
     Enhance
    0.39
    शत
    0.39
    ensä
    0.39
    éon
    0.39
    लेकिन
    0.39
    POSITIVE LOGITS
     balloon
    0.46
     victimes
    0.40
     recentes
    0.40
     superficial
    0.39
     internacionales
    0.39
    гка
    0.38
     intel
    0.38
     filament
    0.38
     glob
    0.37
     victims
    0.37
    Act Density 0.000%

    No Known Activations