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

    Neuronpedia

    INDEX
    Explanations

    prevent some

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ㍍
    0.61
     때문이다
    0.49
    唥
    0.49
    icija
    0.48
    érence
    0.48
    ইন
    0.47
    冖
    0.47
    étera
    0.47
     gamanam
    0.46
     świecie
    0.46
    POSITIVE LOGITS
    0.53
     Quantitative
    0.43
     Clean
    0.43
     Manage
    0.42
     months
    0.41
    acks
    0.40
     associated
    0.40
     name
    0.40
     Use
    0.39
     units
    0.39
    Act Density 0.001%

    No Known Activations