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

    Neuronpedia

    INDEX
    Explanations

    magic

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     bevor
    -0.07
     аналог
    -0.07
     sexuality
    -0.07
     Regina
    -0.06
    ям
    -0.06
     [])↵
    -0.06
    former
    -0.06
     beden
    -0.06
    ayd
    -0.06
    })↵
    -0.06
    POSITIVE LOGITS
     amount
    0.07
     Wood
    0.07
     Tests
    0.07
    sticky
    0.07
    Recipes
    0.07
    Sugar
    0.06
    	length
    0.06
     contents
    0.06
    kening
    0.06
     lemon
    0.06
    Act Density 0.003%

    No Known Activations