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

    Neuronpedia

    INDEX
    Explanations

    first notable instance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     প্রথমে
    0.41
     zaten
    0.39
     sekarang
    0.38
     יותר
    0.38
     তখন
    0.37
     hơn
    0.36
     কয়েক
    0.36
     আছে
    0.36
     আরও
    0.35
     مزید
    0.35
    POSITIVE LOGITS
     ever
    0.76
    ever
    0.72
    本格
    0.66
     truly
    0.63
     véritable
    0.60
     본격
    0.59
     foray
    0.58
    真正
    0.54
    సారి
    0.52
    EVER
    0.51
    Act Density 0.022%

    No Known Activations