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

    Neuronpedia

    INDEX
    Explanations

    most, three, largest

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     allerlei
    0.57
    க்ஸாண்ட
    0.47
     hardening
    0.45
     composants
    0.44
    setType
    0.42
    ாலான
    0.42
    的他
    0.41
    patterns
    0.41
     настрой
    0.41
    לח
    0.41
    POSITIVE LOGITS
     सबसे
    1.18
     three
    1.10
     наиболее
    1.07
     two
    1.05
     most
    1.04
     সবচেয়ে
    1.04
     найбільш
    1.04
     सर्वाधिक
    1.03
     સૌથી
    1.02
     largest
    1.01
    Act Density 0.007%

    No Known Activations