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

    Neuronpedia

    INDEX
    Explanations

    never the answer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    很重要
    0.44
    重要的
    0.41
    ত্যা
    0.40
    关键
    0.39
    重要な
    0.38
    뷸
    0.38
     Gro
    0.38
     crucial
    0.37
     telep
    0.37
     Es
    0.37
    POSITIVE LOGITS
     acceptable
    0.96
     tolerable
    0.84
    acceptable
    0.76
     justifiable
    0.75
     tolerated
    0.69
     justified
    0.68
     permissible
    0.66
     advisable
    0.66
     harmless
    0.63
     excused
    0.62
    Act Density 0.022%

    No Known Activations