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

    Neuronpedia

    INDEX
    Explanations

    sustained pace

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    时代
    0.49
     límite
    0.47
     famílias
    0.47
     chá
    0.47
     વર્ષ
    0.46
    день
    0.46
     yaşad
    0.46
     família
    0.45
     Já
    0.45
     díky
    0.45
    POSITIVE LOGITS
    ibut
    0.46
     honoring
    0.42
     flight
    0.42
     reasonable
    0.42
     effective
    0.41
     culp
    0.41
     reasonably
    0.41
     fairly
    0.40
     critique
    0.40
    (
    0.40
    Act Density 0.001%

    No Known Activations