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

    Neuronpedia

    INDEX
    Explanations

    component or aspect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     N
    0.41
    la
    0.40
     geeign
    0.38
    な
    0.38
     Athens
    0.37
     proches
    0.36
    always
    0.36
     Teenage
    0.36
     A
    0.36
    another
    0.36
    POSITIVE LOGITS
     aspect
    1.26
     part
    1.02
    aspect
    0.91
    部分
    0.90
     aspekt
    0.89
     aspecto
    0.86
     aspects
    0.84
     aspetto
    0.84
     부분을
    0.84
    の部分
    0.83
    Act Density 0.047%

    No Known Activations