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

    Neuronpedia

    INDEX
    Explanations

    subject-verb agreement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    maxWidth
    0.38
     полностью
    0.36
     પોતાના
    0.36
     సన్నివేశ
    0.36
    Anyway
    0.35
     soundness
    0.35
    ልቅ
    0.35
    明顯
    0.35
    दाय
    0.34
     похоже
    0.34
    POSITIVE LOGITS
     verb
    0.43
    had
    0.41
     conjug
    0.39
     might
    0.38
    /
    0.37
     had
    0.37
     did
    0.37
    might
    0.37
     conjugation
    0.35
    ought
    0.35
    Act Density 0.146%

    No Known Activations