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

    Neuronpedia

    INDEX
    Explanations

    specific to a task

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     شامل
    0.41
    łącznie
    0.36
     అల
    0.36
     כול
    0.35
    കൂ
    0.34
    基本的な
    0.34
     chaotic
    0.34
    ToExp
    0.34
    രള
    0.34
    平等
    0.34
    POSITIVE LOGITS
     specific
    2.14
    Specific
    2.03
     Specific
    2.00
    specific
    1.96
     spezif
    1.91
     специфи
    1.85
     específico
    1.84
     específica
    1.83
     specialised
    1.82
     spécifique
    1.82
    Act Density 0.071%

    No Known Activations