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

    Neuronpedia

    INDEX
    Explanations

    pref or settings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    NBC
    0.43
     Purnama
    0.41
     drank
    0.40
     igu
    0.40
     GIRL
    0.40
    獗
    0.39
     EMPC
    0.39
    Drinking
    0.39
     вимо
    0.39
     tube
    0.39
    POSITIVE LOGITS
     Pref
    0.66
    Pref
    0.55
     pref
    0.55
     prefect
    0.54
    pref
    0.49
    ="$
    0.42
        
    0.40
    prefs
    0.39
    settings
    0.37
    sparse
    0.37
    Act Density 0.008%

    No Known Activations