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

    Neuronpedia

    INDEX
    Explanations

    pouring liquids

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Allocator
    -0.07
     seller
    -0.07
    .selection
    -0.07
    	direction
    -0.06
     abych
    -0.06
    ridor
    -0.06
    rella
    -0.06
    814
    -0.06
    	tab
    -0.06
     Killing
    -0.06
    POSITIVE LOGITS
    _SPE
    0.07
     مناس
    0.07
    }));↵
    0.06
    .ed
    0.06
    ('/',
    0.06
    DCF
    0.06
     द
    0.06
     Williamson
    0.06
    �
    0.06
    ичес
    0.06
    Act Density 0.017%

    No Known Activations