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

    Neuronpedia

    INDEX
    Explanations

    installing peft libraries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Cleaning
    0.83
    hemat
    0.79
    nou
    0.78
    Wür
    0.78
    μά
    0.77
    നിക്ക
    0.76
    jär
    0.76
    нков
    0.74
    произ
    0.74
     букмекердик
    0.73
    POSITIVE LOGITS
     Ins
    0.78
     ins
    0.77
     absorbing
    0.72
    얇
    0.69
     absorption
    0.65
    بط
    0.64
     DE
    0.62
     Absorption
    0.61
     behaved
    0.61
     absorb
    0.61
    Act Density 0.003%

    No Known Activations