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

    Neuronpedia

    INDEX
    Explanations

    alternatives or simply

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nejen
    0.22
     WITHOUT
    0.21
     Therefore
    0.17
     Thus
    0.17
     Usually
    0.17
    :
    0.16
     etc
    0.16
     didn
    0.16
     plupart
    0.16
    0.16
    POSITIVE LOGITS
     просто
    0.38
     simplemente
    0.37
     simply
    0.36
     simplesmente
    0.32
    simply
    0.32
     наоборот
    0.32
     alternatively
    0.31
     conversely
    0.30
     addirittura
    0.30
     semplicemente
    0.30
    Act Density 0.331%

    No Known Activations