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

    Neuronpedia

    INDEX
    Explanations

    toxicity and opinion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Over
    0.42
    Integr
    0.42
    Away
    0.41
    Up
    0.41
     startled
    0.41
    Mit
    0.41
    aire
    0.40
    Upp
    0.40
    ála
    0.39
    တို့
    0.39
    POSITIVE LOGITS
     उन्
    0.36
     toxicity
    0.35
     माना
    0.35
    ുന്നത്
    0.34
    rinsic
    0.33
    丁
    0.33
     typical
    0.33
    чении
    0.33
     Likely
    0.33
     toxic
    0.32
    Act Density 0.000%

    No Known Activations