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

    Neuronpedia

    INDEX
    Explanations

    start with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    に使
    0.96
     both
    0.95
    ない
    0.82
     also
    0.81
     quel
    0.81
     supplémentaires
    0.81
    ते
    0.80
     разнови
    0.80
     bakal
    0.80
    度的
    0.79
    POSITIVE LOGITS
    יים
    1.05
     credibility
    0.98
    ar
    0.93
    klar
    0.88
    xref
    0.86
     outset
    0.85
     afresh
    0.84
    ębior
    0.83
    ျေး
    0.82
    ோ
    0.82
    Act Density 0.047%

    No Known Activations