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

    Neuronpedia

    INDEX
    Explanations

    might need or want

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    are
    0.68
     are
    0.64
    os
    0.51
    rett
    0.50
    pes
    0.50
    s
    0.49
    py
    0.49
     is
    0.49
    ruitment
    0.48
    osomal
    0.48
    POSITIVE LOGITS
     conceivably
    0.61
    ާ
    0.59
     mắn
    0.57
     facilmente
    0.57
     offend
    0.57
    あなたの
    0.56
    ב
    0.56
    Ꮈ
    0.55
     diferite
    0.55
    જ
    0.55
    Act Density 0.130%

    No Known Activations