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

    Neuronpedia

    INDEX
    Explanations

    identifying trends and anomalies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     পরিষ
    0.39
     uitvoering
    0.39
     शिक्
    0.37
     savez
    0.37
     인터넷
    0.37
    诠
    0.37
     représ
    0.36
     البعض
    0.36
     dikatakan
    0.36
    emption
    0.35
    POSITIVE LOGITS
     patterns
    0.93
     anomalies
    0.91
     potential
    0.91
     weaknesses
    0.91
     hotspots
    0.88
     flaws
    0.87
     inefficiencies
    0.84
     bottlenecks
    0.84
     vulnerabilities
    0.82
     trends
    0.80
    Act Density 0.047%

    No Known Activations