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

    Neuronpedia

    INDEX
    Explanations

    aligned and unaligned AI

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     últ
    0.53
     ángulos
    0.52
     útiles
    0.51
    olinha
    0.51
    rizes
    0.49
     duquel
    0.49
     árabe
    0.48
     ईरानी
    0.48
    ilih
    0.48
    ASA
    0.47
    POSITIVE LOGITS
     storm
    0.45
     herpes
    0.44
     మొ
    0.43
     neuro
    0.43
    各種
    0.42
    っこ
    0.42
    소를
    0.42
    Copies
    0.42
    参照
    0.41
    Sha
    0.41
    Act Density 0.001%

    No Known Activations