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

    Neuronpedia

    INDEX
    Explanations

    Code and notifications

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (open
    -0.07
    	Init
    -0.06
     Kut
    -0.06
     KT
    -0.06
     More
    -0.06
    Repositories
    -0.06
     favors
    -0.06
     Runtime
    -0.06
    傷
    -0.06
    (Core
    -0.06
    POSITIVE LOGITS
    》的
    0.07
     Barbar
    0.06
    autiful
    0.06
    正确
    0.06
    非常
    0.06
     для
    0.06
    _WAIT
    0.06
    Gratis
    0.06
    ']);↵↵
    0.06
    ERSIST
    0.06
    Act Density 0.005%

    No Known Activations