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

    Neuronpedia

    INDEX
    Explanations

    generate and write content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     downregulation
    0.47
    filtration
    0.44
     harassment
    0.41
    0.41
    และการ
    0.41
     urination
    0.41
     humiliation
    0.41
     playtime
    0.41
     hostility
    0.40
     desperation
    0.40
    POSITIVE LOGITS
     provide
    0.55
     create
    0.52
    provide
    0.50
    create
    0.44
     mitigate
    0.44
     improve
    0.44
     pursue
    0.43
     allocate
    0.43
     establish
    0.42
     encrypt
    0.42
    Act Density 0.755%

    No Known Activations