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

    Neuronpedia

    INDEX
    Explanations

    code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    consuming
    0.43
    María
    0.43
    an
    0.42
    featuring
    0.42
    increasing
    0.41
    sighted
    0.41
    ways
    0.40
     विचार
    0.40
    debate
    0.40
    reducing
    0.40
    POSITIVE LOGITS
     other
    0.41
    newtheorem
    0.40
     altre
    0.40
     autre
    0.39
     Otro
    0.39
     Thrift
    0.38
     Exception
    0.38
     Verdict
    0.38
     Ngoài
    0.38
     Servlet
    0.37
    Act Density 0.000%

    No Known Activations