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

    Neuronpedia

    INDEX
    Explanations

    references to "dragon" in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    dragon

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    =============
    -0.48
    Peek
    -0.37
    因
    -0.34
    ;=
    -0.34
     Meh
    -0.34
     sufrió
    -0.33
    ↵↵↵↵↵↵
    -0.33
     bedient
    -0.33
    Fit
    -0.33
     Midwest
    -0.32
    POSITIVE LOGITS
     Dragon
    2.11
    Dragon
    2.02
     dragon
    1.99
     DRAGON
    1.88
    dragon
    1.77
     Dragons
    1.70
     dragons
    1.70
    Dragons
    1.59
     dragón
    1.48
    dragons
    1.33
    Act Density 0.002%

    No Known Activations