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

    Neuronpedia

    INDEX
    Explanations

    grammar and sentence agreement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     prepayment
    0.42
     sécur
    0.42
     ማ
    0.39
    billing
    0.39
    レール
    0.38
     sinter
    0.38
    學習
    0.38
    duk
    0.38
     motorbike
    0.37
     "~/
    0.37
    POSITIVE LOGITS
     sentences
    0.57
     Present
    0.55
    Present
    0.53
     Presents
    0.50
    sentence
    0.49
     Sentence
    0.47
    Evaluate
    0.45
     sentence
    0.45
    Sentence
    0.45
     cümle
    0.44
    Act Density 0.170%

    No Known Activations