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

    Neuronpedia

    INDEX
    Explanations

    complex systems and abstract concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ولكن
    0.56
    ociaż
    0.56
     oftentimes
    0.49
     اغلب
    0.49
     zá
    0.49
     apprehensive
    0.49
     الف
    0.49
    alten
    0.48
     také
    0.48
    Ĺ
    0.47
    POSITIVE LOGITS
     nontrivial
    0.71
     postdoc
    0.70
     trivially
    0.68
     filesystem
    0.66
     equilibria
    0.63
     metastable
    0.63
     ergodic
    0.63
     bullshit
    0.62
     metast
    0.62
     интернете
    0.61
    Act Density 0.038%

    No Known Activations