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

    Neuronpedia

    INDEX
    Explanations

    testing and sample code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     actual
    0.60
     di
    0.55
     types
    0.55
     examples
    0.54
     experimented
    0.53
     the
    0.52
     example
    0.52
     repeated
    0.52
     de
    0.52
     tests
    0.52
    POSITIVE LOGITS
    ير
    0.49
    SFR
    0.48
    𝙉
    0.46
    Entry
    0.45
     شر
    0.45
     übrig
    0.45
     Polaribacter
    0.45
    0.45
     okam
    0.45
     सुमन
    0.44
    Act Density 0.067%

    No Known Activations