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

    Neuronpedia

    INDEX
    Explanations

    code constructors and keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    return
    0.87
    reaching
    0.76
    reached
    0.74
    break
    0.74
     despe
    0.66
    например
    0.66
    but
    0.65
    suffix
    0.64
    Fmat
    0.64
    newline
    0.64
    POSITIVE LOGITS
     private
    1.10
    ↵↵
    1.09
     Private
    1.01
     constructor
    0.94
    private
    0.93
     privé
    0.91
     prywat
    0.91
     सार्वजनिक
    0.89
     私
    0.87
     Constructor
    0.86
    Act Density 0.052%

    No Known Activations