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

    Neuronpedia

    INDEX
    Explanations

    toxic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     చిత
    0.45
    ের
    0.43
    tage
    0.41
    StillWater
    0.41
    뜨
    0.40
    俬
    0.39
    ात
    0.39
    tis
    0.39
    uum
    0.39
    が現
    0.39
    POSITIVE LOGITS
     toks
    0.98
     Tok
    0.89
     toxic
    0.85
    Tok
    0.84
     tok
    0.80
     toxicity
    0.80
     токси
    0.80
     poisoning
    0.78
     toxins
    0.77
     tox
    0.76
    Act Density 0.016%

    No Known Activations