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

    Neuronpedia

    INDEX
    Explanations

    creating original content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rehabilitation
    0.55
     prep
    0.55
     preparação
    0.53
    現在
    0.52
     préparations
    0.51
     rehabilit
    0.50
     refriger
    0.49
     deeln
    0.47
     rehab
    0.47
     speciali
    0.46
    POSITIVE LOGITS
     Issues
    0.53
    Issues
    0.52
    Create
    0.46
    Creating
    0.45
    Original
    0.44
     Созда
    0.44
     Create
    0.44
    原始
    0.44
     Creating
    0.44
     створи
    0.44
    Act Density 0.001%

    No Known Activations