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

    Neuronpedia

    INDEX
    Explanations

    slows down or becomes problematic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Elon
    0.44
    }))
    0.43
     Slack
    0.43
     יותר
    0.42
     Cv
    0.41
     مؤ
    0.40
     eV
    0.40
     Lift
    0.40
    []
    0.40
    ہر
    0.40
    POSITIVE LOGITS
     retardation
    0.49
    Denote
    0.48
     drawbacks
    0.46
     bý
    0.45
    allocate
    0.45
     empieza
    0.45
     drawback
    0.44
    FloatTensor
    0.44
     imposible
    0.43
     materiale
    0.43
    Act Density 0.006%

    No Known Activations