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

    Neuronpedia

    INDEX
    Explanations

    portion sizes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     replaced
    -0.07
    icolor
    -0.07
    iations
    -0.07
     Interview
    -0.07
    ANA
    -0.07
     epith
    -0.07
     scaling
    -0.07
    ät
    -0.07
    analy
    -0.06
    'ят
    -0.06
    POSITIVE LOGITS
     كبير
    0.06
    .deg
    0.06
    �
    0.06
    　イ
    0.06
    Tipo
    0.06
    remium
    0.06
     podemos
    0.06
    ANCEL
    0.06
    เพลง
    0.06
    fuck
    0.05
    Act Density 0.028%

    No Known Activations