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

    Neuronpedia

    INDEX
    Explanations

    emoji and foreign words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     steam
    1.11
     vapour
    1.05
     condens
    1.00
     vapor
    0.99
     Cond
    0.98
    蒸汽
    0.98
    Steam
    0.95
    steam
    0.95
     Steam
    0.94
     vapeur
    0.92
    POSITIVE LOGITS
         
    1.71
           
    0.98
          
    0.91
    🖕
    0.79
     keer
    0.77
             
    0.77
    🤜
    0.76
     Erwach
    0.76
     ஒத்து
    0.75
    🤙
    0.73
    Act Density 0.375%

    No Known Activations