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

    Neuronpedia

    INDEX
    Explanations

    helpful feedback and suggestions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inescap
    0.47
     inescapable
    0.47
    しなければ
    0.39
     musste
    0.38
    必然
    0.38
     pervasive
    0.37
     relentless
    0.36
     mussten
    0.35
    你应该
    0.34
     orthodox
    0.34
    POSITIVE LOGITS
     helpful
    0.94
     appreciated
    0.85
    helpful
    0.82
     hilfreich
    0.77
     helps
    0.75
    appreciated
    0.75
     Helpful
    0.73
     help
    0.73
     appreciate
    0.69
     hilfre
    0.69
    Act Density 0.006%

    No Known Activations