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

    Neuronpedia

    INDEX
    Explanations

    specifying words, phrases, and language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Método
    0.43
     Fast
    0.40
     multifaceted
    0.38
     massless
    0.37
     tale
    0.37
     humanitarian
    0.36
    VEL
    0.36
     materialism
    0.36
     manuss
    0.35
     EEA
    0.35
    POSITIVE LOGITS
     words
    0.48
     названия
    0.48
     vocabulary
    0.48
    Words
    0.48
     Words
    0.48
     Vocabulary
    0.47
     использу
    0.46
    Vocabulary
    0.46
     phrases
    0.45
    单词
    0.45
    Act Density 0.263%

    No Known Activations