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

    Neuronpedia

    INDEX
    Explanations

    out there / all day

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ૦
    1.02
     cinco
    1.01
     quinze
    0.98
    媺
    0.95
     escoger
    0.95
     турни
    0.89
    ンド
    0.89
    けれど
    0.84
     planejamento
    0.84
     chegando
    0.84
    POSITIVE LOGITS
      
    0.84
    te
    0.78
     Em
    0.74
    ා
    0.73
    itzar
    0.71
    Gü
    0.70
     `
    0.69
    ્સ
    0.69
     overcome
    0.69
     Міні
    0.69
    Act Density 0.000%

    No Known Activations