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

    Neuronpedia

    INDEX
    Explanations

    rm and RMSprop and deletion commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     बार
    0.40
    ECO
    0.39
     hikes
    0.39
     পবিত্র
    0.39
    ಿವ
    0.38
    বাদী
    0.38
     एक्टर
    0.38
     справи
    0.37
     rédaction
    0.37
     jokingly
    0.36
    POSITIVE LOGITS
     rm
    0.64
     RMS
    0.63
     RM
    0.62
     rms
    0.58
    RMS
    0.53
     Rm
    0.43
    RM
    0.42
    rms
    0.42
     danger
    0.41
    Rm
    0.39
    Act Density 0.005%

    No Known Activations