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

    Neuronpedia

    INDEX
    Explanations

    describing qualities and substance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     breeds
    0.86
     breeding
    0.80
     breeders
    0.79
     breeder
    0.77
     BREED
    0.72
    breed
    0.70
     breed
    0.68
     Breeding
    0.66
    breeds
    0.66
     bred
    0.64
    POSITIVE LOGITS
     conformation
    0.62
     conformational
    0.61
     Substance
    0.60
     substance
    0.59
     correct
    0.57
    Correct
    0.57
     correctness
    0.57
    正确
    0.55
    conform
    0.55
     soundness
    0.55
    Act Density 0.015%

    No Known Activations