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

    Neuronpedia

    INDEX
    Explanations

    safe distance, quiet, private location

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ritrov
    0.67
     bilhões
    0.54
     malgré
    0.53
     gigantes
    0.53
     conlleva
    0.53
     महज
    0.53
     relatos
    0.53
     empiezan
    0.52
     aceptas
    0.52
     oure
    0.52
    POSITIVE LOGITS
     preferably
    1.15
    preferably
    1.06
     Preferably
    1.02
     secluded
    1.02
     avoid
    0.90
     isolating
    0.87
     seclusion
    0.86
     isolated
    0.85
     unobstructed
    0.85
    避免
    0.84
    Act Density 0.011%

    No Known Activations