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

    Neuronpedia

    INDEX
    Explanations

    semester academic period

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Tune
    0.44
     overall
    0.44
     protect
    0.43
     Dense
    0.42
     Protect
    0.42
     bucket
    0.41
     Overall
    0.39
     Ent
    0.38
     Eater
    0.38
    bucket
    0.37
    POSITIVE LOGITS
    ttino
    0.45
    jillo
    0.41
     granit
    0.40
    ショナル
    0.39
     POSSIBILITY
    0.39
    shire
    0.38
     جوسینو
    0.38
    ORICAL
    0.38
    splitLength
    0.37
    籶
    0.37
    Act Density 0.000%

    No Known Activations