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

    Neuronpedia

    INDEX
    Explanations

    for

    np_max-act-logits · gemini-2.0-flashTriggered by @bettyfeng2003
    New Auto-Interp
    Negative Logits
    放眼
    -0.07
     Hải
    -0.07
    Phoenix
    -0.07
     avid
    -0.07
    西安市
    -0.06
    的答案
    -0.06
    -in
    -0.06
     Farmers
    -0.06
     glazed
    -0.06
    ,max
    -0.06
    POSITIVE LOGITS
    ::-
    0.07
    _issue
    0.07
    moves
    0.07
    readOnly
    0.07
    _mode
    0.07
    였
    0.07
    T
    0.07
     narrative
    0.07
    뺐
    0.06
    _xt
    0.06
    Act Density 0.003%

    No Known Activations