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

    Neuronpedia

    INDEX
    Explanations

    grammar, spelling, punctuation, word choice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     características
    0.50
     emphas
    0.49
     অনুবাদ
    0.46
    CHANGED
    0.45
     সাহিত্য
    0.45
     emphasizing
    0.45
     alternatif
    0.45
     característica
    0.44
     expressing
    0.44
     storytelling
    0.44
    POSITIVE LOGITS
     misuse
    0.73
     misused
    0.65
     missp
    0.64
     spelling
    0.63
     incorrect
    0.61
     corrigir
    0.60
     improper
    0.60
     illegal
    0.59
     capitalization
    0.59
     missing
    0.58
    Act Density 0.032%

    No Known Activations