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

    Neuronpedia

    INDEX
    Explanations

    thousand

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dimer
    0.55
    十几
    0.54
    においては
    0.54
     Hadamard
    0.53
    <li>
    0.52
     하면
    0.51
     মজা
    0.49
     getchar
    0.49
     ৬০
    0.48
     EntityManager
    0.47
    POSITIVE LOGITS
     thousand
    2.06
     thousands
    1.86
    thousand
    1.82
     Thousand
    1.73
     हजार
    1.70
     тысяч
    1.67
    thousands
    1.66
     Thousands
    1.63
     હજાર
    1.59
    Thousands
    1.58
    Act Density 0.354%

    No Known Activations