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

    Neuronpedia

    INDEX
    Explanations

    grammar, spelling, accuracy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     personajes
    0.36
    ǒ
    0.36
     torches
    0.35
    tout
    0.35
     Trainers
    0.35
    erek
    0.34
     Namen
    0.34
    getVideo
    0.34
     장면
    0.34
     keywords
    0.34
    POSITIVE LOGITS
     accuracy
    0.85
     Accuracy
    0.77
    accuracy
    0.77
    Accuracy
    0.72
     accurate
    0.71
     grammar
    0.71
     fluency
    0.70
     accurately
    0.67
     Grammar
    0.67
     spelling
    0.66
    Act Density 0.019%

    No Known Activations