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

    Neuronpedia

    INDEX
    Explanations

    :

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     cracking
    -0.07
    ocup
    -0.07
    ，她
    -0.06
     Jiang
    -0.06
    ou
    -0.06
    _"+
    -0.06
    와
    -0.06
    ývá
    -0.06
    _PRED
    -0.06
     chiff
    -0.06
    POSITIVE LOGITS
    時に
    0.07
    こと
    0.06
     not
    0.06
    	cs
    0.06
    _sampler
    0.06
    autorelease
    0.06
    */
↵
↵
    0.06
    端
    0.06
    0.06
     autorelease
    0.06
    Act Density 0.010%

    No Known Activations