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

    Neuronpedia

    INDEX
    Explanations

    how do you / how are you

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     looks
    0.52
     влияет
    0.48
     emerges
    0.47
     정말
    0.47
     выглядит
    0.47
     wygląda
    0.46
     evolves
    0.46
     unfolds
    0.46
     evolve
    0.45
     развивается
    0.45
    POSITIVE LOGITS
    define
    0.46
     prefer
    0.45
     Método
    0.41
    justify
    0.41
     Learned
    0.41
     Approach
    0.41
     justify
    0.40
    learned
    0.40
     define
    0.40
     suggest
    0.39
    Act Density 0.021%

    No Known Activations