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

    Neuronpedia

    INDEX
    Explanations

    negativity

    np_max-act · gemini-2.0-flash

    language signaling problems, faults, or negative conditions, often marked by negation or deficiency (e.g., invalidity, lack, inability, mis- states, failures, issues, or harms)

    oai_token-act-pair · gpt-5Triggered by @faiazazmain99

    words with negative connotations indicating problems, defects, wrongness, or undesirable qualities.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @faiazazmain99
    New Auto-Interp
    Negative Logits
    ä¸Ģäºĭ
    -0.28
    ousy
    -0.28
     Saras
    -0.27
    .pub
    -0.27
    licas
    -0.26
    路人
    -0.25
    ertainment
    -0.25
     Shutterstock
    -0.25
    stä
    -0.24
    cased
    -0.24
    POSITIVE LOGITS
    æŶéĹ´ä¸º
    0.29
    ä½ĨçĶ±äºİ
    0.28
    å¿ĥå¾Ĺ
    0.27
    riel
    0.25
     Meg
    0.25
    file
    0.25
    åį´æ²¡æľī
    0.25
    ek
    0.25
    饶
    0.24
    èħº
    0.24
    Act Density 1.623%

    No Known Activations