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

    Neuronpedia

    INDEX
    Explanations

    baseline and baselines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    舧
    0.38
    Inflater
    0.38
     alan
    0.37
    flies
    0.37
     peduncul
    0.37
     Alan
    0.36
    Alan
    0.36
    Tow
    0.36
    돌
    0.35
    tow
    0.35
    POSITIVE LOGITS
    葯
    0.44
     செய்ய
    0.42
    废物
    0.38
     fused
    0.38
    kms
    0.37
    ه‌ها
    0.37
     fuses
    0.37
    स्क
    0.37
    क्या
    0.37
    铑
    0.37
    Act Density 0.000%

    No Known Activations