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

    Neuronpedia

    INDEX
    Explanations

    loss and error minimization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    谦
    0.50
     الخدم
    0.49
     października
    0.47
     أشهر
    0.47
     Świę
    0.47
    市镇
    0.47
     itinéraires
    0.47
    计划
    0.46
     zwölf
    0.46
    ຕິດຕໍ່
    0.46
    POSITIVE LOGITS
     error
    0.72
     loss
    0.72
     minimization
    0.67
     minimized
    0.64
    error
    0.61
     minimize
    0.60
     penal
    0.59
     punishing
    0.57
     penalized
    0.57
     penalty
    0.56
    Act Density 0.157%

    No Known Activations