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

    Neuronpedia

    INDEX
    Explanations

    normally

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    前半
    0.43
     developments
    0.42
     formats
    0.42
     partic
    0.42
     zam
    0.41
     implementation
    0.41
     reinterpret
    0.40
     particularly
    0.40
     گذ
    0.38
     extrem
    0.38
    POSITIVE LOGITS
     ordinarily
    0.75
     normally
    0.71
    normally
    0.70
     normalement
    0.69
     Normally
    0.68
     normalerweise
    0.67
     unthinkable
    0.59
     normalmente
    0.57
    Normally
    0.55
     ordinairement
    0.54
    Act Density 0.072%

    No Known Activations