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

    Neuronpedia

    INDEX
    Explanations

    especially during

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     generalmente
    0.48
     typically
    0.47
    generally
    0.45
     suelen
    0.44
    typically
    0.43
     suele
    0.43
     geralmente
    0.41
    funding
    0.41
     &/
    0.41
     либо
    0.40
    POSITIVE LOGITS
    특히
    0.54
    During
    0.53
     특히
    0.52
    尤其是
    0.51
     DURING
    0.48
    特别是
    0.47
     갤럭시
    0.46
    Especially
    0.46
     آزمایش
    0.46
     במהלך
    0.46
    Act Density 0.018%

    No Known Activations