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

    Neuronpedia

    INDEX
    Explanations

    damage, leaks, and degradation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Cash
    0.76
    ermanfaat
    0.72
     Cash
    0.71
     vyu
    0.70
     montant
    0.70
    简约
    0.69
    ultipl
    0.69
    照顾
    0.69
     menghad
    0.68
    emptyDict
    0.68
    POSITIVE LOGITS
     damage
    1.39
     degradation
    1.38
     cracking
    1.33
     leaks
    1.31
     premature
    1.31
     overheating
    1.30
     deterioration
    1.27
     cracks
    1.24
     excessive
    1.23
     damaging
    1.21
    Act Density 0.352%

    No Known Activations