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

    Neuronpedia

    INDEX
    Explanations

    asking for specifics or thoughts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    必然
    0.42
     needed
    0.41
     требо
    0.39
     necesarios
    0.39
     erforder
    0.39
    plied
    0.39
    に必要な
    0.39
    criptions
    0.38
    needed
    0.38
    要做
    0.38
    POSITIVE LOGITS
     konkrét
    0.75
     specific
    0.73
     интересу
    0.71
     interesse
    0.69
     Interesse
    0.68
     particular
    0.67
     curiosity
    0.67
     aspekt
    0.66
    感兴趣
    0.64
     interests
    0.64
    Act Density 0.014%

    No Known Activations