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

    Neuronpedia

    INDEX
    Explanations

    fundamental building blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     SMTP
    0.89
     bébé
    0.87
    宀
    0.86
     DHCP
    0.86
    า
    0.86
     полости
    0.86
     stroller
    0.85
     doulou
    0.84
     strollers
    0.84
     cough
    0.83
    POSITIVE LOGITS
    hua
    0.89
    𝐺
    0.78
    trk
    0.74
     நல்லது
    0.74
    পালা
    0.73
    𝑙
    0.72
    mAb
    0.72
    valueChanged
    0.72
    됨
    0.71
    ptr
    0.70
    Act Density 0.155%

    No Known Activations