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

    Neuronpedia

    INDEX
    Explanations

    describing inclusions and components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     formando
    0.36
     explicando
    0.34
     можете
    0.34
     forming
    0.34
     Contribute
    0.33
     formar
    0.33
     Simply
    0.32
     Used
    0.31
    只会
    0.31
     ayudan
    0.31
    POSITIVE LOGITS
     incorporates
    1.11
     includes
    1.09
     involves
    1.03
     include
    0.98
     involve
    0.97
     incluye
    0.96
     incorporate
    0.93
    includes
    0.92
     включает
    0.89
    incorpor
    0.89
    Act Density 0.096%

    No Known Activations