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

    Neuronpedia

    INDEX
    Explanations

    refuses harmful request

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    di
    0.40
    การ
    0.39
     Pereira
    0.39
    gunaan
    0.39
     bras
    0.39
    राट
    0.38
    Theorem
    0.38
     सीई
    0.38
     Showcase
    0.38
     таксама
    0.38
    POSITIVE LOGITS
     fulfils
    0.46
     //#
    0.43
     expands
    0.42
     expanded
    0.42
     perplexed
    0.41
     বড়ই
    0.41
    钕
    0.40
    尼亚
    0.40
    ","#
    0.40
     fulfills
    0.39
    Act Density 0.000%

    No Known Activations