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

    Neuronpedia

    INDEX
    Explanations

    promote, foster, develop

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     keeps
    0.75
    线性
    0.74
     Keeps
    0.73
     endlessly
    0.70
     incroyable
    0.70
     rápidamente
    0.68
     Staying
    0.68
     کہاں
    0.68
    өлү
    0.68
     کہنے
    0.67
    POSITIVE LOGITS
     promote
    1.73
    Prom
    1.66
    promote
    1.61
     Promote
    1.56
     foster
    1.55
     promoting
    1.55
     promotion
    1.52
     promoted
    1.49
    promotion
    1.45
    促进
    1.44
    Act Density 0.037%

    No Known Activations